WEBVTT
00:00:00.320 --> 00:00:02.959
Welcome to the Everyday AI podcast.
00:00:03.120 --> 00:00:09.439
My name is Jordan Wilson, and for the past three and a half years, we put out more than 800 episodes.
00:00:09.599 --> 00:00:13.759
Yet one of the most common questions I get, I didn't really have an answer for.
00:00:14.000 --> 00:00:16.960
Where do I start on the Everyday AI podcast?
00:00:17.280 --> 00:00:19.760
And that's why we started the Start Here series.
00:00:19.920 --> 00:00:30.239
And with Fall now back in full swing, the Everyday AI podcast is going back to school and playing back the entire Start Here series from front to back.
00:00:30.399 --> 00:00:37.759
We've hit pause on our normal Monday to Friday programming to run back our most popular series ever for the next 30 days.
00:00:37.920 --> 00:00:42.000
We made the Start Here series for beginners and AI champions alike.
00:00:42.159 --> 00:00:53.359
So whether you're just trying to get a grasp on large language models or grappling with the best coding harness for multi-agentic workflows, the Start Here series covers it all.
00:00:53.600 --> 00:00:58.079
Plain language, no jargon, and easy to follow along each day.
00:00:58.320 --> 00:01:03.679
So make sure to subscribe to the podcast and check back each day for new insights day by day.
00:01:03.840 --> 00:01:11.040
The series is a culmination of spending more than 10,000 hours covering generative AI over the past three and a half years.
00:01:11.200 --> 00:01:14.719
So you don't want to miss a single episode of the Start Here series.
00:01:14.879 --> 00:01:16.159
Let's get into it.
00:01:18.000 --> 00:01:22.079
Maybe you aren't fully agent-pilled like me yet.
00:01:22.319 --> 00:01:23.599
And that's okay.
00:01:23.840 --> 00:01:31.760
It could be because your company hasn't given you the green light to go full desktop agent yet with Codex or Claude Code.
00:01:32.000 --> 00:01:39.840
Or maybe you're still just getting comfortable with the web interfaces of Google Gemini, ChatGPT, or Anthropics Claude.
00:01:40.000 --> 00:01:50.239
But don't worry, you can still get a taste of that sweet, proactive, agentic life where you start your morning and a good chunk of your work is already done for you.
00:01:50.480 --> 00:01:51.599
Well, how?
00:01:52.079 --> 00:01:56.799
If you can't be using these autonomous desktop agents, how in the world can you get that?
00:01:56.959 --> 00:01:59.680
Well, they're called scheduled tasks.
00:01:59.840 --> 00:02:07.599
So think of scheduled tasks in today's large language models kind of as the gateway drug to being fully agent-pilled.
00:02:07.840 --> 00:02:21.280
So if you haven't or can't make the big desktop leap to running agents around the clock, you're not out of luck because these scheduled tasks are a legit secret weapon.
00:02:21.439 --> 00:02:29.439
So secret, in fact, that sometimes it's hard to even find it and the capabilities aren't even fully known or documented.
00:02:29.680 --> 00:02:36.159
So we're going to be diving into that on today's show as we put AI to work on Wednesdays.
00:02:36.319 --> 00:02:41.199
Yeah, it is our day where we go hands-on, but here is the big picture.
00:02:41.439 --> 00:02:49.599
Scheduling is one of the most underrated features in today's large language models, at least when it comes to being on the web.
00:02:49.759 --> 00:02:58.879
Because most of your routine use cases that you're using large language models for right now, well, you're probably repetitive and you're probably doing a lot of the same things over and over.
00:02:58.960 --> 00:03:01.039
And well, they can just be done for you.
00:03:01.120 --> 00:03:11.280
Instead of you just being a button pusher doing the same type of tasks, whether you know it or not, you can actually just have those be done automatically and wake up and have them done.
00:03:11.439 --> 00:03:20.960
Uh, but I think scheduling tasks has gone through so many different name and interface changes in all the platforms, it's sometimes a little confusing, or you might not know how it works.
00:03:21.120 --> 00:03:28.960
And there's actually a handful of hidden features when you're scheduling inside of specifically Chat GPT and Claude.
00:03:29.199 --> 00:03:32.240
So on today's show, that's exactly what we're gonna dive into.
00:03:32.319 --> 00:03:40.240
And here's what you're gonna learn you're gonna learn the basics on why scheduling is so important and how to easily do it in the main platforms.
00:03:40.400 --> 00:03:44.639
You're gonna know what features work well in scheduling and which ones don't.
00:03:44.800 --> 00:03:53.120
And I'm gonna tell you the best use cases for scheduling to take a step toward a more proactive and agentic workflow.
00:03:53.439 --> 00:03:55.120
All right, let's get into it.
00:03:55.199 --> 00:03:55.680
What's going on?
00:03:55.759 --> 00:03:56.719
My name is Jordan Wilson.
00:03:56.800 --> 00:03:58.000
Welcome to Everyday AI.
00:03:58.159 --> 00:04:00.240
If you're new here, we do this every single day.
00:04:00.319 --> 00:04:10.639
It's your unedited, unscripted live stream podcast and free daily newsletter, helping business leaders like you and me keep up with all the feature updates, the advancements, the UI changes, how the heck do you do this?
00:04:10.800 --> 00:04:11.840
What's going on?
00:04:12.000 --> 00:04:13.680
I do all that work for you.
00:04:13.840 --> 00:04:18.240
I tell you what works, what doesn't, what you need to pay attention to to grow your company and your career.
00:04:18.399 --> 00:04:20.160
So if that's what you're trying to do, sweet.
00:04:20.319 --> 00:04:22.160
Starts here with the live stream and the podcast.
00:04:22.240 --> 00:04:26.720
But if you haven't already, please make sure to go to our website at your everydayai.com.
00:04:26.800 --> 00:04:34.399
We're gonna be recapping not just the highlights from today's show, but everything else you need to know, especially the other AI news and developments, to stay ahead.
00:04:34.639 --> 00:04:38.240
And hey, actually a quick little announcement.
00:04:38.399 --> 00:04:40.319
It is back to school time.
00:04:40.399 --> 00:04:41.519
Yeah, here in the US.
00:04:41.600 --> 00:04:45.120
It's that time of the year where everyone goes back to school.
00:04:45.279 --> 00:04:51.040
But uh, maybe you're not a student or maybe you don't even have kids in school, but guess what?
00:04:51.199 --> 00:04:52.879
We're all going back to school together.
00:04:53.040 --> 00:04:57.600
Uh, so we're gonna be doing some learning together on everyday AI.
00:04:57.759 --> 00:05:07.439
So there's actually this thing that we've been doing for a while called our Start Here series that we kicked off earlier this year, and they've been some of our most popular shows ever.
00:05:07.680 --> 00:05:15.279
It covers all of the basics of not just you know yesterday's AI, but today's AI and where it's going.
00:05:15.360 --> 00:05:21.600
Uh, so we're gonna be running back the entire series starting this Monday with the first volume.
00:05:21.759 --> 00:05:36.079
So a little break from our normal Monday to Friday scheduling, but I guarantee if you stick with this series from volume one through volume 30 and you put the work in, you'll know more than about 99% of the other people in your company when it comes to AI.
00:05:36.160 --> 00:05:40.240
So no jargon, no fluff, no computer science degree required.
00:05:40.480 --> 00:05:43.199
Class starts Monday, don't be late.
00:05:43.360 --> 00:05:52.800
All right, let's get into today's episode, though, where we're gonna be talking about scheduling AI and how to easily make AI work for you, right?
00:05:52.879 --> 00:05:55.439
In Claude Gemini in Chat GPT.
00:05:55.519 --> 00:05:57.120
Yeah, because I don't know.
00:05:57.519 --> 00:06:00.319
I think there was a certain time before this was a feature.
00:06:00.480 --> 00:06:02.720
I was like, can we get this, please?
00:06:02.800 --> 00:06:03.040
Right.
00:06:03.120 --> 00:06:13.600
I remember seeing back in like uh 2024 when there was rumors, right, of these kind of scheduled tasks or you know, scheduled actions, they've gone under a lot of name changes.
00:06:13.759 --> 00:06:20.639
I remember when these were just leaks and rumors, and I'm like, oh my gosh, this will completely change the way the world uses large language models.
00:06:20.720 --> 00:06:21.680
And guess what?
00:06:21.920 --> 00:06:22.879
It didn't.
00:06:23.199 --> 00:06:33.439
Mainly because I think a lot of the companies kind of changed things around so much, or they buried these and made them hard to find, which is actually um interesting.
00:06:33.839 --> 00:06:50.160
Because I think that the uh kind of the agentic adoption was slower than it maybe could have been, if not if if these uh you know kind of scheduled options or features would have been earlier in front and center.
00:06:50.240 --> 00:06:55.199
Uh, because if I'm being honest, I was a very early adopter of these.
00:06:55.279 --> 00:06:56.480
Like the day it came out.
00:06:56.639 --> 00:06:59.680
I remember I think it was ChatGPT tasks that was the first to come out.
00:06:59.839 --> 00:07:01.839
It might have been in like late 2024.
00:07:02.000 --> 00:07:05.920
I started moving all of my workflows over to ChatGPT tasks.
00:07:06.160 --> 00:07:12.000
And I think that process, uh, you know, I always talk about this concept of unlearning and rebuilding.
00:07:12.160 --> 00:07:26.000
I think that was actually one of the most important steps for me to prepare myself for, you know, moving most of our processes and even my thinking, my creativity, my brainstorming into an agentic workflow.
00:07:26.079 --> 00:07:36.879
Because, you know, you really have to start deconstruct, uh, you know, to deconstruct your entire process when you're like, okay, wait, I can wake up and have all of this stuff done for me, right?
00:07:37.040 --> 00:07:42.240
You start, I think, uh thinking less in terms of what do I need to do next.
00:07:42.399 --> 00:07:45.199
Instead, you start thinking, what is the end goal?
00:07:45.279 --> 00:07:45.519
Right.
00:07:45.600 --> 00:07:51.519
And then you start working your way backwards from that and saying, how can I have this end goal waiting for me in the morning?
00:07:51.759 --> 00:07:52.319
All right.
00:07:52.800 --> 00:07:56.399
So uh let's go through the three main platforms.
00:07:56.560 --> 00:07:56.879
All right.
00:07:57.040 --> 00:08:03.680
And you know, if you don't know, the three main platforms, anyways, are uh Gemini, Claude, and Chat GPT.
00:08:03.759 --> 00:08:08.800
So yeah, technically, Copilot, it's just too messy to do this, right?
00:08:08.879 --> 00:08:16.319
I can show you the online version of Copilot, but almost every single one of our listeners, if you're using Copilot, you're using it on the desktop.
00:08:16.399 --> 00:08:31.439
The reason why I'm not doing Copilot, well, number one, I'm on a Mac, uh, but number two, uh, even if I was on a PC, it is entirely too difficult in terms of permissions, in terms of jumping through all of these different hoops.
00:08:31.600 --> 00:08:39.759
So it's almost like uh to show you these things, it would kind of be a waste because you would go out and be like, oh my gosh, this is amazing, and you would try and do it, and it wouldn't work.
00:08:39.919 --> 00:08:46.159
I actually did a show previously on the online version of Microsoft tasks, which I think are really good.
00:08:46.320 --> 00:08:49.279
So uh, you know, I'll make sure to drop that one in today's newsletter.
00:08:49.360 --> 00:08:51.840
But let's get into uh Google Gemini.
00:08:52.159 --> 00:08:59.840
So here's the quick rundown on uh Gemini's version, and these are called Gemini's Scheduled Actions.
00:09:00.080 --> 00:09:03.039
So let me just be honest, not that good.
00:09:03.200 --> 00:09:10.080
They're hard to find, they're buggy, it's extremely limited in terms of features and the user interface.
00:09:10.240 --> 00:09:13.440
And you also can't schedule via normal chat.
00:09:13.600 --> 00:09:18.320
And unfortunately, when you try to do that, you're actually gonna get some false positives.
00:09:18.639 --> 00:09:31.519
So the other downside is if you once you do get them scheduled, um, if you don't use them or check in on them, uh they will be automatically disabled, which stinks, right?
00:09:31.759 --> 00:09:33.840
Um, so yeah, a lot of downsides.
00:09:34.000 --> 00:09:36.159
So here's how you're going to find them.
00:09:36.240 --> 00:09:40.399
So if you just go to gemini.google.com, make sure you're logged in.
00:09:40.480 --> 00:09:43.600
So if you're looking on the left-hand panel, they're not gonna be there.
00:09:43.840 --> 00:09:44.159
All right.
00:09:44.320 --> 00:09:56.080
So, you know, fortunately, uh Chad GPT and Claude did uh move these uh scheduled tasks for them, that's what they're called, front and center, on the main nav bar.
00:09:56.240 --> 00:09:58.720
So uh Google Gemini, no such luck.
00:09:58.879 --> 00:10:03.519
So you have to actually look in your settings, and there you'll see these scheduled actions.
00:10:03.679 --> 00:10:04.159
All right.
00:10:04.480 --> 00:10:16.399
So once you do that, um, then essentially you are met with a very simple and straightforward enough um scheduled action box where you just essentially type it in.
00:10:16.559 --> 00:10:22.159
So you give it a name, you give it instructions, you say how often, and then a time.
00:10:22.399 --> 00:10:24.559
And that's it, right?
00:10:24.639 --> 00:10:35.440
It's an extremely simple interface, no bells, no whistles, and the downside is well, it's not that useful, if I'm being honest.
00:10:35.600 --> 00:10:36.000
All right.
00:10:36.159 --> 00:10:44.480
Uh, so next up, and I am gonna, I do have an example of you know, one kind of simple S uh prompt that I did.
00:10:44.639 --> 00:10:59.120
Um, and then at the very end, after I show you these examples, so yeah, this is one FYI and this is one is technically pre-recorded, uh, because it's gonna be it was a little difficult to, you know, go through uh all of my different systems without accidentally doxing myself.
00:10:59.279 --> 00:11:01.279
And also these are scheduled, right?
00:11:01.360 --> 00:11:07.519
So I didn't want to schedule these for, you know, oh, 745 and then oh, I'm done before then, right?
00:11:07.600 --> 00:11:10.639
Because you know, the live stream kicks off at 7:30 a.m.
00:11:10.799 --> 00:11:11.600
Central Standard time.
00:11:11.759 --> 00:11:16.000
Uh so it was gonna be a little tricky to try to get all these to schedule live and in real time.
00:11:16.159 --> 00:11:22.320
So yeah, this is one of those shows for AI at work on Wednesdays, normally very hands-on, doing live prompts, all that.
00:11:22.399 --> 00:11:28.159
So this is more of uh, you know, hey, I'm gonna show you how to cook, put it in the oven, and guess what?
00:11:28.240 --> 00:11:29.840
It's already done, got the cake ready.
00:11:30.000 --> 00:11:34.559
All right, so next up, let's talk about how ChatGPT's scheduling works.
00:11:34.799 --> 00:11:37.759
So theirs are called scheduled tasks.
00:11:37.919 --> 00:11:44.080
And again, I actually did a dedicated show uh on uh ChatGPT's scheduled tasks.
00:11:44.159 --> 00:11:51.600
Let me see what uh what one that was, because it was actually fairly recent because they went through some major changes.
00:11:51.759 --> 00:11:56.480
All right, so this was actually on July 1st, so episode 810.
00:11:56.639 --> 00:12:02.559
So I'm gonna go over the high levels, uh high level points for that, but it was actually a pretty in-depth episode.
00:12:02.639 --> 00:12:04.559
I remember I spent a good amount of time on it.
00:12:04.639 --> 00:12:10.799
Uh, so if you are a heavy chat GPT user, like a lot of people are, uh, make sure you go check out episode 810.
00:12:10.879 --> 00:12:12.159
But here's the gist.
00:12:12.399 --> 00:12:17.519
So tasks now support more agenc app actions, and that's the great thing.
00:12:17.679 --> 00:12:24.080
They also work inside of projects, and you can use memory for more personalized automation.
00:12:24.240 --> 00:12:25.519
That part's huge.
00:12:25.679 --> 00:12:38.240
Uh so you can access them directly from the sidebar, like I talked about before, or you can go to chatgbt.com/slash scheduled, and you can just create one by describing what you want ChatGPT to do in natural language.
00:12:38.480 --> 00:12:42.639
Like, you know, you can say scan my Gmail daily at 7 a.m.
00:12:42.799 --> 00:12:44.480
and tell me what's important.
00:12:45.759 --> 00:12:52.320
All right, but there's a lot more to uncover here because this is where ChatGPT really stands out.
00:12:52.480 --> 00:13:02.240
So a little um kind of hack, uh, you can go through the normal uh interface to set this up, or like I said, you can set it up conversationally.
00:13:02.320 --> 00:13:13.600
But what a lot of people don't understand is if you set this up conversationally, you actually have um some options that you don't have if you were to set it up in your normal scheduling routine.
00:13:13.759 --> 00:13:18.720
Uh, that's because, and I'm kind of looking on my on my other screen uh right now.
00:13:18.879 --> 00:13:32.559
So if you go into the scheduled section, um, you're gonna see it's it's gonna look like a fairly uh a fairly normal prompt box inside of Chat GPT, except there's one or two things missing.
00:13:32.639 --> 00:13:36.320
And the big thing here is the model selector.
00:13:36.480 --> 00:13:46.320
So yeah, it's um, I believe what happens in the schedule task, it just is gonna default to the last model you use, which may or may not be a good thing.
00:13:46.559 --> 00:14:00.559
So I recommend um just doing scheduling a task via natural language in the main kind of new chat box and just say schedule this at this time and chat GPT is smart enough where it's gonna do that.
00:14:00.720 --> 00:14:05.519
The main benefit there is well, you can choose the exact model you want.
00:14:05.679 --> 00:14:09.039
So maybe you want to use, you know, GPT-5.
00:14:10.240 --> 00:14:12.639
Soul extra high, right?
00:14:12.879 --> 00:14:17.919
At least you have the ability to configure the exact level of reasoning that you need for something.
00:14:18.000 --> 00:14:20.879
Uh, because maybe it's a little simpler, right?
00:14:21.039 --> 00:14:27.200
Maybe it's more of a personalized search and you don't exactly need the most, you know, the smartest model.
00:14:27.279 --> 00:14:37.200
But if I'm being honest, it's one of those things where uh unless you're running out of limits on a ChatGPT Pro plan, which is pretty much impossible.
00:14:37.360 --> 00:14:44.320
Um, it's you might as well just use the most powerful model at this because it's gonna be scheduled, it's gonna be happening while you sleep.
00:14:44.480 --> 00:14:46.879
So throw as much juice as you can at these things.
00:14:46.960 --> 00:14:51.200
Uh so that's a little tip there with the scheduling tasks inside ChatGPT.
00:14:51.360 --> 00:15:01.039
Uh, the other one, a little hack, and for whatever reason, this doesn't always work because maybe it is truly a hack.
00:15:01.279 --> 00:15:18.960
But if you go into the new work tab inside of Chat GPT, okay, uh, and I know that there's a lot of confusion about this, so I'm not gonna dive in too deeply, but essentially there's a new version of the Chat GPT desktop, which is called ChatGPT Work, which is kind of the same thing as codex.
00:15:19.200 --> 00:15:26.960
But then if you go to chat GPT.com, you'll see at the top now there's a new chat in a new uh work section.
00:15:27.120 --> 00:15:33.759
So the easiest way to tell the difference, because people are like, wait, when I go to chat gpt.com, do I use chat or do I use work?
00:15:34.000 --> 00:15:39.440
Well, the simplest way is well, use chat when you just want to chat or talk to Chat GPT.
00:15:39.600 --> 00:15:42.559
But if you want something more like an agent, use work.
00:15:43.120 --> 00:15:51.600
So the cool thing here, and this is again, I've never seen or heard anyone else doing this, and it doesn't work every single time.
00:15:51.679 --> 00:15:55.519
But if you get it to work once when scheduling your task, for me, it works over and over.
00:15:55.679 --> 00:16:04.080
But the cool thing in the work mode, it actually has its own virtual and dedicated computer and browser, which is awesome, right?
00:16:04.159 --> 00:16:11.759
So if there's anything that you normally have to do that requires a ton of interface clicking around, right?
00:16:11.919 --> 00:16:16.159
It could be from uh a website that you rely on every single day.
00:16:16.320 --> 00:16:22.159
It could just be a lot of you know web searching that you can't just do from reading a static page, right?
00:16:22.320 --> 00:16:32.080
Because a lot of times when these large language models, you know, quote unquote use the web, they essentially just scan the code um of any given page, which is helpful in most cases.
00:16:32.159 --> 00:16:35.679
But what if you have to, you know, scroll down to load more?
00:16:35.840 --> 00:16:41.360
Uh, what if you have to um you know click around a bunch and add filters, right?
00:16:41.440 --> 00:16:47.440
So think of all the web browsing that you do that maybe you haven't yet handed off um to an agent.
00:16:47.519 --> 00:16:49.440
And there's probably a reason for that.
00:16:49.679 --> 00:16:51.440
And this is awesome.
00:16:51.600 --> 00:16:54.480
So, you know, one quick example of something like this.
00:16:54.960 --> 00:16:57.039
I love Google Trends, right?
00:16:57.120 --> 00:17:01.840
So there's something in Google Trends, you go, you can put all these filters on, and then you have to scroll down.
00:17:01.919 --> 00:17:05.519
I like to see like what's trending in big tech in AI.
00:17:05.759 --> 00:17:08.960
And I realized, well, why am I doing that?
00:17:09.119 --> 00:17:09.359
Right.
00:17:09.440 --> 00:17:40.319
So that's something I went into Chat GPT, clicked the work, uh, did just a natural language, gave it the URL to that, you know, once I put all those filters on uh the Google trend, and then I said, Hey, every day at 9 p.m., go to this URL, scroll down to the bottom, uh, see, you know, use everything that you know about me and the everyday AI audience and see which of these trends that our audience would care the most about, then go out and research all those trends, tell me why they're trending, right?
00:17:40.480 --> 00:17:45.359
Like uh yesterday something about the iPhone was trending and I didn't know why, right?
00:17:45.440 --> 00:17:55.279
So it's gonna go and it's gonna, you know, help me stay more in touch with things that matter to me in my career without me having to go through and click and then go out and research.
00:17:55.359 --> 00:17:57.519
I just wake up and I know, right?
00:17:57.599 --> 00:18:03.279
So that's actually something inside of Chad GPT that is very unique to it because uh, you know, Claude can't do that.
00:18:03.440 --> 00:18:06.720
Uh, you know, we're not talking co-pilot, co-pilot can't do that.
00:18:06.799 --> 00:18:09.759
And obviously, Gemini's scheduled actions can't do that.
00:18:09.839 --> 00:18:10.960
So big hack there.
00:18:11.119 --> 00:18:16.079
Um, another one that's very unique to Chad GPT is having these open-ended tasks.
00:18:16.160 --> 00:18:19.759
So this is something that uh OpenAI didn't even really highlight.
00:18:19.839 --> 00:18:22.960
Um, in I think it was technically late June that this came out.
00:18:23.039 --> 00:18:26.160
We covered it like the next day, I think it was like July 1st, right?
00:18:26.559 --> 00:18:32.799
Um, but you can now just say something like, you know, check my competitor's pricing page and tell me when it changes.
00:18:32.960 --> 00:18:35.920
You don't have to say check it every Tuesday or at this PM.
00:18:36.400 --> 00:18:40.640
I don't know how this magic works, but it just does, right?
00:18:40.799 --> 00:18:42.000
Uh so that's really cool.
00:18:42.079 --> 00:18:44.559
You can also just give it open-ended tasks.
00:18:45.359 --> 00:18:46.000
All right.
00:18:46.400 --> 00:18:49.200
Um, so now I'm showing on my screen how that works.
00:18:49.279 --> 00:18:56.480
So on the left-hand side, you have your um uh scheduled tasks there.
00:18:56.640 --> 00:19:07.039
Uh, then uh you can go in to the scheduled task, and like I said, you still get uh you know, you can use all your connectors, your skills, all of those drop down.
00:19:07.119 --> 00:19:20.240
But the downside is you don't have that model selector, but you can still create a scheduled task from the scheduled section, or you can just click new chat and just type it out um in natural language, give a time, a frequency, etc.
00:19:20.799 --> 00:19:23.440
Um, then you can go in and edit them.
00:19:23.519 --> 00:19:32.240
So if you when you do click that uh scheduled portion, or if you go to chat gpt.com slash scheduled, uh, you can go in and edit it.
00:19:32.319 --> 00:19:39.279
So if you do give a prompt in natural language, it's gonna turn that into a set of better instructions for the model to follow.
00:19:39.440 --> 00:19:42.000
But then from there, you can make any edits.
00:19:42.079 --> 00:19:50.880
Uh so if you remember the kind of old school uh GPT builder, you could just chat with ChatGPT and it would turn your chats into a set of custom instructions.
00:19:51.039 --> 00:19:53.279
Kind of the same thing happens here under the hood.
00:19:53.359 --> 00:19:57.839
So you can go in and update the end uh scheduled task.
00:19:57.920 --> 00:20:05.839
And you can also, there's a drop-down menu for when you want it to repeat, the frequency, the time, and an end date.
00:20:06.799 --> 00:20:09.839
All right, let's keep going and talk about clauds.
00:20:10.000 --> 00:20:14.480
So clauds are powerful, but a little confusing.
00:20:14.640 --> 00:20:16.240
Okay, and here's why.
00:20:16.400 --> 00:20:19.519
Uh, because you have your normal scheduled tasks.
00:20:19.599 --> 00:20:22.480
So again, this is all just Claude on the web.
00:20:22.720 --> 00:20:25.680
Okay, so I'm not talking, you know, codex desktop.
00:20:25.759 --> 00:20:30.000
I'm not talking about, you know, Claude Desktop with Claude Code and Claude Cowork.
00:20:30.079 --> 00:20:31.839
Uh, I am talking about Claude on the web.
00:20:32.079 --> 00:20:39.279
But if you didn't know this, Claude on the web also changed a little bit, just kind of like how ChatGPT did, how you have chat and work.
00:20:39.359 --> 00:20:42.799
Uh, so now Claude has home in code.
00:20:42.960 --> 00:20:48.400
All right, so if you go to home, uh you can do scheduled tasks that way.
00:20:48.480 --> 00:20:55.440
So you essentially can do a scheduled task in normal chat or in Claude cohort, even more confusing.
00:20:55.519 --> 00:20:58.319
So there's technically like three different ways you can set this up.
00:20:58.480 --> 00:21:02.880
Uh, but uh the other way is a routine.
00:21:03.039 --> 00:21:07.839
So if you click on the Claude Code button, you can set up a routine.
00:21:07.920 --> 00:21:11.680
And I'm gonna quickly tell you uh the difference here in the pros and the cons.
00:21:11.839 --> 00:21:22.480
For for Claude's scheduled tasks, um, that's just the uh it is very much the equivalent of the ChatGPT version, which is like chat and work, right?
00:21:22.559 --> 00:21:27.440
So if you need something that's just more straightforward, you know, just use the the home.
00:21:27.519 --> 00:21:32.400
Uh, if you need something that's a little more technically uh agentic, you can click the co-work tab.
00:21:32.559 --> 00:21:41.200
But you can also, just like you can in ChatGPT, you can create a scheduled task just conversationally with Claude, or there is a manual uh setup.
00:21:41.359 --> 00:21:44.480
Uh, the cool thing about Clauds is you can work in a project.
00:21:44.640 --> 00:21:56.640
You can also set an approval level per task, because um, as an example, all right, if if you're giving it send actions for the first time it runs, you might want to approve those.
00:21:56.720 --> 00:21:59.759
Uh, I don't fully understand uh if you're trying.
00:22:00.160 --> 00:22:01.200
To schedule, right?
00:22:01.279 --> 00:22:09.519
I kind of scratched my head on this because I'm like, okay, why would you actually want uh you know to like manually approve everything if it's on a schedule, right?
00:22:09.599 --> 00:22:14.640
Like if you have 10 of these things running overnight, why would you not rent them auto-approved?
00:22:14.720 --> 00:22:17.440
So I guess in theory, it's a nice feature.
00:22:17.680 --> 00:22:20.240
I just literally don't know the point of it.
00:22:20.400 --> 00:22:21.920
But, anyways, it's there.
00:22:22.079 --> 00:22:27.359
Maybe if you just want to run it once in front of you to see how it works, maybe that's what it's there for.
00:22:27.519 --> 00:22:36.559
Um, otherwise, it's extremely counterintuitive to schedule something, but then oh, it's not gonna fully run because you had to approve it.
00:22:36.799 --> 00:22:42.480
Anyways, uh, you can also choose the model, but you cannot unfortunately choose the reasoning level.
00:22:42.640 --> 00:22:55.839
You can set the frequency, it doesn't have a custom uh version like ChatGPT, but you can also uh use connectors and skills like you can in Chat GPT, but then you have these scheduled routines.
00:22:55.920 --> 00:22:58.160
And I actually did this, let me see.
00:22:58.240 --> 00:23:02.319
This this show was a little uh a little longer ago.
00:23:02.480 --> 00:23:03.680
Let me see when I did this.
00:23:03.839 --> 00:23:06.319
The scheduled uh routines.
00:23:06.480 --> 00:23:10.880
Yeah, this is live all or you know, unedited, unscripted at uh at least.
00:23:11.119 --> 00:23:12.880
So let's see when I did this show.
00:23:13.039 --> 00:23:20.079
Um so this was okay, this was episode 756.
00:23:20.640 --> 00:23:34.319
Um this was in April, but these are all uh still very, very relevant because um, you know, we went over this was technically through the scope of Claude Desktop, but it kind of worked the exact same.
00:23:34.480 --> 00:23:41.839
But we talked in depth about the differences uh between these scheduled tasks and the routines in Claude Code.
00:23:41.920 --> 00:23:48.400
But the difference on the routine, um, right, is well, you can actually sync those with your desktop.
00:23:48.480 --> 00:23:56.960
Uh so if you set up a Claude Schedule task, it's not gonna sync inside Claude um desktop.
00:23:57.039 --> 00:24:03.279
So the only way you can do it is with a Claude Code routine, and then you uh have to choose the cloud version.
00:24:03.440 --> 00:24:11.839
So um on codex, uh right on the desktop version, uh, if you're using ChatGPT work to schedule, it'll all sync.
00:24:12.079 --> 00:24:15.519
Uh so why would you use Claude's scheduled routines?
00:24:15.599 --> 00:24:20.079
And I'll just give you the 30-second version of uh episode 756.
00:24:20.240 --> 00:24:21.519
I think they're really, really good.
00:24:21.680 --> 00:24:31.519
Um, they're more powerful, and I think a lot of people, especially non-technical people, are gonna look at like Claude routines and be like, uh, Claude code and be like, I'm not touching those, right?
00:24:32.400 --> 00:24:37.279
You're gonna say there's enough with skills and all these other things and and co-work, right?
00:24:37.359 --> 00:24:38.720
That's technical enough for me.
00:24:38.960 --> 00:24:41.759
The claw claude routines are super simple.
00:24:41.920 --> 00:24:47.599
The you know, you you don't even really have to use clawed code to go in and use claud routines.
00:24:47.839 --> 00:24:57.599
Um, the the the biggest um kind of benefit is you can use um essentially an API token as a trigger.
00:24:57.759 --> 00:25:00.160
Uh, so what that means, it's very simple.
00:25:00.240 --> 00:25:05.680
Uh, you can choose that, and I have a screenshot of that later so I can uh walk through.
00:25:06.079 --> 00:25:07.359
Oh, I have it right here.
00:25:07.519 --> 00:25:14.000
Um, so essentially, if you choose that as an option, you're gonna get essentially an API token.
00:25:14.160 --> 00:25:16.640
All right, and then you can go to something like Zapier.
00:25:16.880 --> 00:25:27.759
Um, and you can use that to essentially connect anything in Zapier and have that be a trigger to start something in a Claude routine, which is really cool, right?
00:25:27.839 --> 00:25:42.960
Because as an example, you know, in ChatGPT, they actually used to have this, I believe, in Codecs, but then they kind of simplified and unified uh the the codecs in Chat GPT uh they used to be called automations, so now they're just called schedule tasks.
00:25:43.119 --> 00:25:45.119
Anyways, this used to be available in codecs.
00:25:45.200 --> 00:25:49.200
I don't believe it is anymore, but this is great in uh claude routines.
00:25:49.440 --> 00:26:00.480
So, as an example, you can set anything uh that happens in Zapier as a trigger to start um to start a routine inside of Claude code.
00:26:00.640 --> 00:26:02.480
Again, you don't have to be super technical.
00:26:02.640 --> 00:26:05.839
You can literally just give it to Claude and be like, walk me through this.
00:26:05.920 --> 00:26:13.599
Um, right, but uh that's super valuable because otherwise, every other thing has to start at a certain time, right?
00:26:13.759 --> 00:26:23.839
So maybe I don't know, you you have a super uh you know, um highly valuable uh form on your company's website, right?
00:26:23.920 --> 00:26:30.720
Uh when someone comes in, it's a potential million-dollar client, and you want to be able to start doing some legwork right away when it comes in.
00:26:30.880 --> 00:26:34.720
Obviously, you can't do a schedule task on that because you have no clue.
00:26:34.880 --> 00:26:42.960
But in this case, right, you can set this up inside Claude Code, takes like 30 seconds, choose that as an option, you're gonna get this key.
00:26:43.200 --> 00:26:49.039
Then you go into Zapier, you know, whatever it is for using a Google form, Jot form, whatever.
00:26:49.119 --> 00:26:52.000
Uh, you know, then you're gonna get this key, plug it into Zapier.
00:26:52.160 --> 00:27:00.400
And then anytime that web form comes in, then that routine is going to start inside of Claude Code once it kind of sends to that API token.
00:27:00.559 --> 00:27:06.000
So uh a nice unique feature there inside the uh scheduled routines.
00:27:06.559 --> 00:27:09.519
All right, so uh that's kind of the gist of it.
00:27:09.680 --> 00:27:15.839
And I'm gonna go through now and uh I'm gonna share my screen for our live stream audience and podcast audience.
00:27:15.920 --> 00:27:20.240
You can always go to our website at your everydayai.com if you want to see the video version of this.
00:27:20.400 --> 00:27:23.920
Nothing super visual, and I'll kind of explain what's going on.
00:27:24.079 --> 00:27:24.880
Um, right.
00:27:24.960 --> 00:27:31.440
And or you can always check the show notes and we uh do uh put a link directly to today's episode.
00:27:31.599 --> 00:27:35.759
All right, so uh let's talk about what our test actually was.
00:27:35.920 --> 00:27:38.240
So this one was simple-ish.
00:27:39.039 --> 00:27:39.839
Um, right.
00:27:40.319 --> 00:27:50.799
The downside is because Google Gemini uh is not good at even using its own connectors like Gmail.
00:27:50.880 --> 00:28:01.039
Um, I was trying to find something that I thought in theory uh could be a very um, you know, apples to apples comparison and something maybe realistic.
00:28:01.200 --> 00:28:04.480
And also this is something that I wanted to build, anyways.
00:28:04.640 --> 00:28:10.480
Um, so I said, all right, let's just throw a simple schedule prompt um and talk about the use case of this.
00:28:10.720 --> 00:28:15.039
So um, you know, I said every day at blank o'clock, right?
00:28:15.119 --> 00:28:18.799
Uh, because I spaced these out a little bit, but I gave it the same thing.
00:28:18.880 --> 00:28:28.079
Uh so I said every day at this time, please put together an interactive either site or artifact or canvas, uh, because they all call them something different.
00:28:28.160 --> 00:28:32.799
But this is essentially an interactive visual, and in some it worked and some it didn't, and I'll show you that.
00:28:32.960 --> 00:28:52.319
Um, so at every day at uh a certain time, please put together an interactive visual that shows the current stock prices for these AI stocks NVIDIA, Alphabet, Microsoft, Broadcom, Meta Platforms, uh, Taiwan Semiconductor, Micron Technologies, uh, AMD, Oracle, Palantir.
00:28:52.640 --> 00:29:01.039
Uh, also, please include an up-to-date factual bullet point recap for each that explains the biggest news for each of these companies.
00:29:01.279 --> 00:29:13.759
So, this is something that I don't do it every single day, but I probably every single day end up checking at least two to three um of these companies in terms of what's happening with their stock.
00:29:14.000 --> 00:29:27.359
Not because I, you know, have stock in all of them and I'm tracking it from a portfolio perspective, but from an AI news perspective, a lot of times a company stock will shoot up five percent or drop five percent, which is very rare.
00:29:27.440 --> 00:29:31.759
And a lot of times it has to do with something that's happening in AI, as an example.
00:29:31.920 --> 00:29:50.480
You know, Alphabet slash, you know, the parent company of Google uh announced a major uh AI leadership change um, you know, a couple of days ago, and their stock dropped like five percent, which for a multi-trillion dollar market cap company is an insane amount of money, right?
00:29:50.559 --> 00:29:55.519
So sometimes I just want to be able to see at a glance what's happening with all these big AI companies.
00:29:55.599 --> 00:29:59.839
And you know, essentially if I see a 3% up or down, that's kind of my threshold.
00:30:00.160 --> 00:30:05.359
Then I know that something meaningful is happening probably in AI around news.
00:30:05.519 --> 00:30:07.839
So this is something I kind of wanted to build for myself.
00:30:08.000 --> 00:30:10.960
But the thing is, well, it refreshes all the time.
00:30:11.039 --> 00:30:20.640
So that's the good thing with this being a scheduled right, this is now technically an agent that's going out and doing this every single day for me and updating it so I don't have to.
00:30:20.880 --> 00:30:26.480
All right, so let's go ahead and look at the results.
00:30:27.119 --> 00:30:28.960
Let's share my screen here.
00:30:29.200 --> 00:30:34.799
All right, so Google Gemini first and Google Gemini kind of failed.
00:30:34.960 --> 00:30:36.960
All right, so what's interesting?
00:30:37.119 --> 00:30:42.480
So I did say use canvas mode, um, but it didn't.
00:30:42.799 --> 00:31:03.200
So um I do believe I could be mistaken, but I do believe when you used to have uh the um scheduled actions um here inside of uh Google Gemini, I thought you used to be able to use some of the other features, specifically uh the canvas mode.
00:31:03.599 --> 00:31:06.319
Uh, but in this case, it didn't work.
00:31:06.559 --> 00:31:08.720
Um, so that's interesting.
00:31:08.880 --> 00:31:11.519
It did, however, go out and do the rest.
00:31:11.680 --> 00:31:14.240
So it actually fired at the correct time.
00:31:14.400 --> 00:31:16.799
So I did this one at 5 15 p.m., right?
00:31:16.880 --> 00:31:18.559
So after the markets closed.
00:31:18.720 --> 00:31:24.880
Um, so it got the correct prices for NVIDIA, and then it gave me the key news, alphabet, Microsoft, etc.
00:31:25.279 --> 00:31:28.079
So it did fire at the correct time.
00:31:28.240 --> 00:31:39.119
It followed instructions, um, it uh, you know, got the correct prices, but it didn't put together a visual uh that I wanted to use in Canvas mode, right?
00:31:39.200 --> 00:31:50.160
I could go down here um and I could say, right, I'll actually do this just so we can actually just make it at least give Google and Gemini a chance because otherwise I'm just ragging on Gemini this entire time.
00:31:50.319 --> 00:31:53.839
So I'm gonna say, uh, you know, using all this info.
00:31:54.799 --> 00:31:55.119
All right.
00:31:55.279 --> 00:32:00.480
So yeah, unfortunately, I'm gonna have to um put together a two-step one on this.
00:32:00.640 --> 00:32:09.839
So uh this is where I also wanted to, you know, test out and say, hey, what other features um are available when you're scheduling a task?
00:32:10.000 --> 00:32:15.200
Because sometimes uh uh features or modes make all the difference.
00:32:15.359 --> 00:32:20.079
It's the difference between, you know, using your data and not using your data, uh, right?
00:32:20.160 --> 00:32:30.160
Like spoiler alert, one of the reasons I'll I've never really used uh Google Google Gemini's uh scheduled tasks is because they don't use your connected data, right?
00:32:30.240 --> 00:32:32.720
So it's like for me, it's not super useful.
00:32:32.880 --> 00:32:37.599
And in this case, it wasn't able to use the mode um for Canvas.
00:32:37.759 --> 00:32:41.359
All right, next, let's go ahead and take a look.
00:32:42.400 --> 00:32:44.319
Uh let's take a look at Claude.
00:32:44.880 --> 00:32:46.559
So uh same thing.
00:32:46.720 --> 00:32:51.119
Um, I did tell it to use an artifact.
00:32:51.359 --> 00:32:53.920
It however didn't.
00:32:54.079 --> 00:32:59.839
So it just wrote some strict HTML, but I don't know.
00:32:59.920 --> 00:33:03.839
Um, I'm not gonna say it's a failure because it looks good enough, uh, right?
00:33:04.000 --> 00:33:10.240
It actually put together a uh very clean uh, you know, it says AI stocks daily.
00:33:10.319 --> 00:33:11.680
It's interactive, right?
00:33:11.759 --> 00:33:14.319
It has a light mode, a dark mode.
00:33:14.400 --> 00:33:20.720
Uh, it gave me very clear ticker symbols, uh, the percentage that they're up and down on the day, right?
00:33:20.799 --> 00:33:22.799
Like here I see Google is down 3%.
00:33:23.440 --> 00:33:26.720
So I can click there and see what happened, right?
00:33:26.799 --> 00:33:34.400
So it said, you know, Tuesday it fell uh, you know, 3.8% amid a rotation of high valuation mega cap tech.
00:33:34.720 --> 00:33:38.799
It says investors remain uneasy about AI spending, right?
00:33:38.880 --> 00:33:45.440
As their cap X guidance was uh about 200 billion, uh, and their capex was up 100% year over year.
00:33:45.680 --> 00:33:49.440
So this for me, this is all really helpful stuff to read, right?
00:33:49.680 --> 00:33:54.000
So when I'm talking about AI things, I you know can ground it in a little more recent data.
00:33:54.160 --> 00:33:59.920
Uh, there's also this button that it says, you know, uh expand all news, collapse all news, which is really nice.
00:34:00.079 --> 00:34:04.240
So it didn't technically make it like a shareable artifact.
00:34:04.400 --> 00:34:09.199
Um, but overall, you know, it it did it did pretty good.
00:34:09.360 --> 00:34:15.519
Um so uh Gemini, you know, probably failure.
00:34:15.679 --> 00:34:20.480
Uh you know, uh Claude got a passing grade, but not ideal.
00:34:20.719 --> 00:34:23.599
But let's take a look at Chat GPT.
00:34:23.840 --> 00:34:26.639
So my gosh, this one's good.
00:34:26.719 --> 00:34:27.039
All right.
00:34:27.119 --> 00:34:36.960
So yeah, if if if you're not looking um at the live stream here, this is one of those ones where you know, sometimes I'm like, oh yeah, it's cool.
00:34:37.039 --> 00:34:38.559
This one's really good.
00:34:38.800 --> 00:34:39.920
This one's really good.
00:34:40.079 --> 00:34:43.199
All right, for a zero shot prompt, right?
00:34:43.280 --> 00:34:44.719
I read it to you, super simple.
00:34:44.960 --> 00:34:47.280
Um did a fantastic job.
00:34:47.360 --> 00:34:51.519
So I used ChatGPT sites, which by the way is a freaking cheat code.
00:34:51.599 --> 00:34:55.039
Like, if you're not using ChatGPT sites, it is in ChatGPT now.
00:34:55.119 --> 00:34:59.679
It used to only be available um inside of uh codecs.
00:34:59.840 --> 00:35:02.880
It used to only be available for team plans.
00:35:03.039 --> 00:35:07.599
Now it's available even on the web and it's available on individual paid plans as well.
00:35:07.760 --> 00:35:12.079
If you're not using ChatGPT sites, I'm just saying this, you're literally missing out.
00:35:12.400 --> 00:35:14.639
I make them all the time, right?
00:35:14.719 --> 00:35:20.480
I have custom skills uh to make a lot of them in kind of a templated way that really work for me.
00:35:20.639 --> 00:35:22.800
Uh, I send them to my wife, right?
00:35:22.880 --> 00:35:24.079
Because you can make them public.
00:35:24.239 --> 00:35:28.800
Uh, anyways, let's take a look at what uh Chat GPT sites did.
00:35:29.360 --> 00:35:33.760
Uh first of all, visually, fantastic, right?
00:35:33.920 --> 00:35:35.119
It gave me this breakdown.
00:35:35.360 --> 00:35:38.480
You know, it says four of the 10 advanced, six of the 10 declined.
00:35:38.559 --> 00:35:41.920
It gave me what's called an AI basket pulse, which is pretty cool.
00:35:42.079 --> 00:35:45.920
So it said overall, it was down 0.6% across 10 companies.
00:35:46.000 --> 00:35:49.360
So I can see overall, you know, how are these things moving?
00:35:49.679 --> 00:35:52.000
Give me the winners, the losers, uh, right?
00:35:52.079 --> 00:35:54.639
I can click on the biggest moves, the winners first.
00:35:54.719 --> 00:35:57.840
Uh, I can create a watch list, uh, which is pretty cool, right?
00:35:57.920 --> 00:36:00.239
I can toggle this little star on and off.
00:36:00.320 --> 00:36:09.119
Um, and then when I click on any of these, um it then brings the uh all of the news that mattered.
00:36:09.199 --> 00:36:10.079
So let's find another.
00:36:10.159 --> 00:36:11.440
We already talked about Google.
00:36:11.679 --> 00:36:17.280
Or no, actually, let's let's click Google and see if it gave me similar uh reasons that Google was down almost 4%.
00:36:17.679 --> 00:36:23.199
So it says Alphabet reported Q2 revenue of 120 billion, up 24 year over year.
00:36:23.280 --> 00:36:36.079
Search search and other revenue rose 17, uh Google Cloud revenue grew 82%, and backlog reached 514 billion with management attributing the acceleration to AI infrastructure in AI solutions.
00:36:36.159 --> 00:36:38.000
Uh, so I will say this.
00:36:38.320 --> 00:36:41.519
Uh, it pulled some older news, right?
00:36:41.679 --> 00:36:48.320
So uh if I were to redo this, uh obviously the ChatGPT one did the best, uh, but it pulled in some older news.
00:36:48.480 --> 00:36:57.920
Uh, I don't know if the Google one didn't give me the date of the news, um, but I'm guessing this news was maybe a little more recent.
00:36:58.079 --> 00:36:58.960
I'm not sure.
00:36:59.119 --> 00:37:05.679
Uh, but from a visual uh UI UX perspective, it's just much better, right?
00:37:05.840 --> 00:37:13.679
I'm actually gonna go through and uh, you know, probably make one or two changes in this and you know, make it mobile optimized.
00:37:13.760 --> 00:37:18.880
And the good thing is, is then all I do, you know, on my phone, I can essentially save this as an app.
00:37:18.960 --> 00:37:26.320
And then, you know, I can save it to my home screen, you know, and every day when I open it, it's gonna reload and bring up the freshest information.
00:37:26.559 --> 00:37:36.159
So um yeah, overall, uh I think they all had their own strengths, right?
00:37:36.239 --> 00:37:37.360
Uh, let's take a look.
00:37:37.440 --> 00:37:38.719
Um, going back over.
00:37:39.840 --> 00:37:43.199
Uh, sure enough, uh, we had another error.
00:37:43.360 --> 00:37:46.320
Um, another error in uh Google Gemini.
00:37:46.480 --> 00:37:46.800
That's fine.
00:37:46.880 --> 00:37:51.760
I really try to give uh Gemini in Canvas mode, uh, which I've always loved their canvas mode.
00:37:51.840 --> 00:37:52.480
I think it's really good.
00:37:52.559 --> 00:37:53.119
There we go.
00:37:53.199 --> 00:37:54.960
Uh so now it's working.
00:37:56.079 --> 00:37:57.039
Is it okay?
00:37:57.199 --> 00:37:57.679
Interesting.
00:37:57.840 --> 00:37:59.519
So it went a completely different direction.
00:37:59.679 --> 00:38:04.079
Give me kind of bar charts, uh, you know, prices, quarter two growth, capex.
00:38:04.480 --> 00:38:06.960
Uh so you know, it actually looks nice.
00:38:07.039 --> 00:38:11.920
Uh, some nice features and functionalities, not exactly what I was expecting.
00:38:12.000 --> 00:38:16.079
Uh, but then okay, when I scroll down, uh, I think it does a pretty good job.
00:38:16.159 --> 00:38:17.440
Let's read Google.
00:38:17.679 --> 00:38:22.480
Um, so it says uh executive recap summary.
00:38:22.719 --> 00:38:26.639
Although the Google it's not right, right?
00:38:26.719 --> 00:38:34.239
Uh, because it says it's it's up one percent, uh almost one percent for the day, which um I don't believe it was right now.
00:38:34.400 --> 00:38:37.280
I now I have to double check uh to make sure.
00:38:37.440 --> 00:38:39.440
Yeah, Google was down almost four percent.
00:38:39.599 --> 00:38:46.960
So Google was wrong about the actual uh day, so that kind of ruins it, right?
00:38:47.039 --> 00:38:53.760
So I'm not even gonna look at the rest because if you get the facts wrong, right, you always gotta you always gotta check these things, right?
00:38:53.840 --> 00:39:03.840
But yeah, uh Google was actually down you know, three point, you know, roughly seven percent, which the other uh models got correct, and Google unfortunately got it wrong.
00:39:03.920 --> 00:39:10.079
So even though you know it did finish the dashboard with multiple prompts, get a fail, Google.
00:39:10.159 --> 00:39:10.880
We get a fail.
00:39:10.960 --> 00:39:11.360
All right.
00:39:11.599 --> 00:39:15.440
So let's wrap up here and talk about my takeaways.
00:39:15.599 --> 00:39:18.239
So overall, Gemini, not useful at all.
00:39:18.400 --> 00:39:20.400
It struggles using its own connectors.
00:39:20.559 --> 00:39:27.599
Uh essentially, the only thing that it's really useful for is to customize web search for you.
00:39:27.760 --> 00:39:33.440
Uh, but even like you saw here, when in my again, my one example, it pulled in incorrect information.
00:39:33.679 --> 00:39:36.719
So uh I haven't used Google's, right?
00:39:36.800 --> 00:39:42.559
Aside from when I test these things, I haven't used Google's um scheduled actions.
00:39:42.639 --> 00:39:43.920
I hope they improve them.
00:39:44.000 --> 00:39:53.440
Uh, right, we'll see uh when uh we get a whether it's a Gemini 3.7 Pro, Gemini 3.5 Pro, Gemini 4 Pro, whatever.
00:39:53.519 --> 00:39:56.719
Uh hopefully they bring some updates to the Gemini web platform.
00:39:56.800 --> 00:39:58.639
They did go through a UI UX update.
00:39:58.719 --> 00:40:02.480
So hopefully these scheduled actions uh get some love as well.
00:40:02.559 --> 00:40:14.400
But overall, I'm not using them and I would say you probably shouldn't either, unless Gemini is literally the only thing that you can use at work, and the only thing you really need is like a customized Google search.
00:40:14.480 --> 00:40:16.719
Um, because otherwise, really not useful.
00:40:16.800 --> 00:40:17.280
Uh, all right.
00:40:17.360 --> 00:40:20.719
My main Chad GPT takeaway is it's really good.
00:40:20.880 --> 00:40:25.519
There's a lot of unique benefits, like the you know, the any-time scheduling, right?
00:40:25.599 --> 00:40:30.880
Uh just hey, go check when this happens and it somehow figures it out without you having to schedule it.
00:40:31.119 --> 00:40:35.679
Um, using work mode with the virtual computer and browser is really good.
00:40:36.000 --> 00:40:40.559
Um, the custom schedule the others don't have, like if you say like every other Tuesday, right?
00:40:40.639 --> 00:40:44.000
You can't do that um in Claude or in Gemini.
00:40:44.320 --> 00:40:54.719
Uh having an end date is a unique feature as well in ChatGPT because maybe as an example, you want to run it every single day for two weeks, but then you don't want it to keep running and clog up your feet.
00:40:54.880 --> 00:40:55.679
Uh, it's great.
00:40:55.840 --> 00:41:07.599
I think ChatGPT, uh, their schedule tasks more than any others are great at contextualizing um your history or contextualizing your schedule task in your data and history, right?
00:41:07.760 --> 00:41:11.760
I forgot the technical uh update for ChatGPT's memory.
00:41:11.840 --> 00:41:14.559
I think it's called like Dream V3 or something.
00:41:14.800 --> 00:41:20.079
It is so much better, so much better than you know anyone else out there.
00:41:20.159 --> 00:41:21.519
It's not even close.
00:41:21.679 --> 00:41:24.800
Um, I again don't know how OpenAI does it.
00:41:24.960 --> 00:41:28.079
Um, but that actually really comes into play.
00:41:28.159 --> 00:41:30.400
Um, when you're scheduling these things, right?
00:41:30.559 --> 00:41:38.719
Especially when they're of high value, you want to be able to bring in all of that other context and history from everything that ChatGPT or Claude knows about you.
00:41:38.880 --> 00:41:50.079
So unfortunately, right, like if you're doing the Claude routine, uh, which is really helpful, Claude Routine has no clue what you've been doing in Claude Chat or in Claude Cowork.
00:41:50.239 --> 00:41:54.159
So you can't wrap it in that extremely important business context.
00:41:54.320 --> 00:41:59.360
All right, ChatGPT, you can also use your connectors and plugins and skills, which is really cool.
00:41:59.599 --> 00:42:04.239
Um then, like I said, being able to tap into the full power of ChatGPT.
00:42:04.320 --> 00:42:10.719
So both in the reasoning levels is extremely important and helpful, or being able to use the sites feature, right?
00:42:10.800 --> 00:42:13.599
Having that sites feature that just updates all the time, amazing.
00:42:13.760 --> 00:42:19.280
All right, so now let's talk about Claude, the home or the chat scheduling takeaway.
00:42:19.440 --> 00:42:22.559
So more straightforward, but it works, right?
00:42:22.719 --> 00:42:26.000
Unlike Google, you can use all your connectors, which is good.
00:42:26.159 --> 00:42:29.119
Being able to set the permissions, even though I don't fully understand that.
00:42:29.199 --> 00:42:32.079
If you want to run it once as a test, maybe that's helpful.
00:42:32.320 --> 00:42:45.039
Overall, straightforward, nothing, you know, there's no standout features in the um kind of home or co-work scheduling, but there is um in the other version in the routines.
00:42:45.119 --> 00:42:51.360
So in the routines, there's a lot of, you know, I already went over the um kind of the two, well, two unique features.
00:42:51.440 --> 00:42:52.400
I didn't mention the other one.
00:42:52.480 --> 00:43:04.239
The one is being able to kind of have that API token um and have that act as a trigger, which is very unique in terms of what you can do to schedule right now inside of Claude Code routines.
00:43:04.400 --> 00:43:06.719
Like I said, don't be scared of Claude Code.
00:43:06.800 --> 00:43:10.239
You don't even have to use anything else in Claude Code, uh, right?
00:43:10.400 --> 00:43:17.599
You can just think of it like a normal chat, but you can kick off um a routine based on literally anything that happens on the web.
00:43:17.679 --> 00:43:29.920
Just go in Zapier, you know, take that API token and say anything that happens in the 6,000 platforms that Zapier um connects with, send it to this API token, and then you can kick off any routine.
00:43:30.079 --> 00:43:36.239
Uh, a great um a great use case and something unique for all of the others when it comes to scheduling your work.
00:43:36.400 --> 00:43:48.639
And then the other one, which is a small one, um, inside clawed um claud code routines, you can also set routine level messaging alerts, which sounds like okay, why does it matter?
00:43:48.800 --> 00:43:55.599
Um, you know, in ChatGPT and uh Gemini, you can set notifications for all of them, but not at the individual level.
00:43:55.679 --> 00:44:05.119
So one of the reasons why that's helpful, uh, right, is you can, as an example for a certain routine that runs, you can send a Slack message, uh, you know, which is pretty helpful.
00:44:05.440 --> 00:44:05.840
All right.
00:44:06.000 --> 00:44:08.719
So, like I said, there's pros and there's cons.
00:44:08.880 --> 00:44:16.559
I think overall, um, Chat GPT is the winner with a close second to the routines scheduling of Claude.
00:44:16.719 --> 00:44:23.599
I wish Claude would just be able to unify those somehow um and bring some of those better features in routines into their schedule tasks.
00:44:23.679 --> 00:44:26.639
But uh, hey, if you split it up, you can still get the power.
00:44:26.719 --> 00:44:32.639
But yeah, unfortunately, it doesn't kind of have that cohesive memory, which I think is so important to wrap your scheduled tasks up in.
00:44:32.719 --> 00:44:37.360
Otherwise, you may spend more time going back and adjusting them and making them better, anyways.
00:44:37.679 --> 00:44:40.880
All right, so what should you schedule as we wrap here?
00:44:41.039 --> 00:44:45.760
Well, I would ask codecs or chat GPT work first, right?
00:44:45.840 --> 00:45:01.199
Uh unfortunately, Claude Code doesn't have cross-chat or cross-cowork code memory, um, but uh unless it commits something to memory, but I would go through and spend time seeing what you actually do manually inside of these platforms.
00:45:01.360 --> 00:45:03.599
From there, find those repetitive tasks.
00:45:03.679 --> 00:45:05.119
That's what you should be scheduling.
00:45:05.280 --> 00:45:18.480
The other thing, kind of a little hack here, is look at your browser history or give it to an agent and see what your most viewed websites are, and then tie those websites to a deliverable and then schedule a task for that, right?
00:45:18.559 --> 00:45:28.400
The biggest thing is being able to do these things that are time intensive, whether it's research for the Chat GPT work hack that I told you, clicking around on a bunch of websites, right?
00:45:28.559 --> 00:45:39.360
See where you're spending your most time, either manually pushing the buttons in Gemini, in Chat GPT, in Claude, turn those into scheduled tasks, or just where are you spending a lot of your time on the internet?
00:45:39.599 --> 00:45:40.880
The internet's a wasteful place.
00:45:40.960 --> 00:45:42.159
Don't spend time there.
00:45:42.320 --> 00:45:47.280
All right, and then last but not least, the other one is start with this simple triage dashboard, right?
00:45:47.360 --> 00:45:49.360
That's updated every single day.
00:45:49.519 --> 00:46:11.280
It's obviously only going to work in Chat GPT and Claude, uh, but you know, bringing in your you know connectors for mail, calendar, Slack, Teams, your CRM, your major KPIs, your most used connectors, MCPs, etc., being able to automatically have an agent go through, pull all the most important things, and essentially have a start here dashboard, right?
00:46:11.360 --> 00:46:13.760
It's a very, you know, simple use case.
00:46:13.920 --> 00:46:21.039
Maybe that's the 2026 version of like the 2024, uh, you know, improve this email, right?
00:46:21.199 --> 00:46:27.199
Probably in 2026, it's essentially having a custom dashboard and triaging everything and telling you where to start.
00:46:27.440 --> 00:46:33.840
But if you are newer to the game and you haven't already done that, I think that's a great place to start with a scheduled task or action.
00:46:34.079 --> 00:46:37.119
All right, that is a wrap, y'all.
00:46:37.199 --> 00:46:46.159
Uh, so remember, if you haven't already, we're gonna be kicking off going back to school with our start here series uh starting Monday.
00:46:46.239 --> 00:46:47.199
So don't be late.
00:46:47.280 --> 00:46:48.159
I'll see you then.
00:46:48.320 --> 00:46:53.760
And if you haven't already, please go to your everydayai.com, sign up for the free daily newsletter.
00:46:53.840 --> 00:46:54.480
Thanks for tuning in.
00:46:54.559 --> 00:46:57.199
We'll see you back tomorrow in every day for more everyday AI.
00:46:57.280 --> 00:46:58.159
Thanks, y'all.
00:46:58.960 --> 00:47:02.239
And that's a wrap for today's edition of Everyday AI.
00:47:02.400 --> 00:47:03.519
Thanks for joining us.
00:47:03.760 --> 00:47:07.519
If you enjoyed this episode, please subscribe and leave us a rating.
00:47:07.599 --> 00:47:09.199
It helps keep us going.
00:47:09.440 --> 00:47:16.559
For a little more AI magic, visit your everydayai.com and sign up to our daily newsletter so you don't get left behind.
00:47:16.719 --> 00:47:19.360
Go break some barriers, and we'll see you next time.
00:00:00.320 --> 00:00:02.959
Welcome to the Everyday AI podcast.
00:00:03.120 --> 00:00:09.439
My name is Jordan Wilson, and for the past three and a half years, we put out more than 800 episodes.
00:00:09.599 --> 00:00:13.759
Yet one of the most common questions I get, I didn't really have an answer for.
00:00:14.000 --> 00:00:16.960
Where do I start on the Everyday AI podcast?
00:00:17.280 --> 00:00:19.760
And that's why we started the Start Here series.
00:00:19.920 --> 00:00:30.239
And with Fall now back in full swing, the Everyday AI podcast is going back to school and playing back the entire Start Here series from front to back.
00:00:30.399 --> 00:00:37.759
We've hit pause on our normal Monday to Friday programming to run back our most popular series ever for the next 30 days.
00:00:37.920 --> 00:00:42.000
We made the Start Here series for beginners and AI champions alike.
00:00:42.159 --> 00:00:53.359
So whether you're just trying to get a grasp on large language models or grappling with the best coding harness for multi-agentic workflows, the Start Here series covers it all.
00:00:53.600 --> 00:00:58.079
Plain language, no jargon, and easy to follow along each day.
00:00:58.320 --> 00:01:03.679
So make sure to subscribe to the podcast and check back each day for new insights day by day.
00:01:03.840 --> 00:01:11.040
The series is a culmination of spending more than 10,000 hours covering generative AI over the past three and a half years.
00:01:11.200 --> 00:01:14.719
So you don't want to miss a single episode of the Start Here series.
00:01:14.879 --> 00:01:16.159
Let's get into it.
00:01:18.000 --> 00:01:22.079
Maybe you aren't fully agent-pilled like me yet.
00:01:22.319 --> 00:01:23.599
And that's okay.
00:01:23.840 --> 00:01:31.760
It could be because your company hasn't given you the green light to go full desktop agent yet with Codex or Claude Code.
00:01:32.000 --> 00:01:39.840
Or maybe you're still just getting comfortable with the web interfaces of Google Gemini, ChatGPT, or Anthropics Claude.
00:01:40.000 --> 00:01:50.239
But don't worry, you can still get a taste of that sweet, proactive, agentic life where you start your morning and a good chunk of your work is already done for you.
00:01:50.480 --> 00:01:51.599
Well, how?
00:01:52.079 --> 00:01:56.799
If you can't be using these autonomous desktop agents, how in the world can you get that?
00:01:56.959 --> 00:01:59.680
Well, they're called scheduled tasks.
00:01:59.840 --> 00:02:07.599
So think of scheduled tasks in today's large language models kind of as the gateway drug to being fully agent-pilled.
00:02:07.840 --> 00:02:21.280
So if you haven't or can't make the big desktop leap to running agents around the clock, you're not out of luck because these scheduled tasks are a legit secret weapon.
00:02:21.439 --> 00:02:29.439
So secret, in fact, that sometimes it's hard to even find it and the capabilities aren't even fully known or documented.
00:02:29.680 --> 00:02:36.159
So we're going to be diving into that on today's show as we put AI to work on Wednesdays.
00:02:36.319 --> 00:02:41.199
Yeah, it is our day where we go hands-on, but here is the big picture.
00:02:41.439 --> 00:02:49.599
Scheduling is one of the most underrated features in today's large language models, at least when it comes to being on the web.
00:02:49.759 --> 00:02:58.879
Because most of your routine use cases that you're using large language models for right now, well, you're probably repetitive and you're probably doing a lot of the same things over and over.
00:02:58.960 --> 00:03:01.039
And well, they can just be done for you.
00:03:01.120 --> 00:03:11.280
Instead of you just being a button pusher doing the same type of tasks, whether you know it or not, you can actually just have those be done automatically and wake up and have them done.
00:03:11.439 --> 00:03:20.960
Uh, but I think scheduling tasks has gone through so many different name and interface changes in all the platforms, it's sometimes a little confusing, or you might not know how it works.
00:03:21.120 --> 00:03:28.960
And there's actually a handful of hidden features when you're scheduling inside of specifically Chat GPT and Claude.
00:03:29.199 --> 00:03:32.240
So on today's show, that's exactly what we're gonna dive into.
00:03:32.319 --> 00:03:40.240
And here's what you're gonna learn you're gonna learn the basics on why scheduling is so important and how to easily do it in the main platforms.
00:03:40.400 --> 00:03:44.639
You're gonna know what features work well in scheduling and which ones don't.
00:03:44.800 --> 00:03:53.120
And I'm gonna tell you the best use cases for scheduling to take a step toward a more proactive and agentic workflow.
00:03:53.439 --> 00:03:55.120
All right, let's get into it.
00:03:55.199 --> 00:03:55.680
What's going on?
00:03:55.759 --> 00:03:56.719
My name is Jordan Wilson.
00:03:56.800 --> 00:03:58.000
Welcome to Everyday AI.
00:03:58.159 --> 00:04:00.240
If you're new here, we do this every single day.
00:04:00.319 --> 00:04:10.639
It's your unedited, unscripted live stream podcast and free daily newsletter, helping business leaders like you and me keep up with all the feature updates, the advancements, the UI changes, how the heck do you do this?
00:04:10.800 --> 00:04:11.840
What's going on?
00:04:12.000 --> 00:04:13.680
I do all that work for you.
00:04:13.840 --> 00:04:18.240
I tell you what works, what doesn't, what you need to pay attention to to grow your company and your career.
00:04:18.399 --> 00:04:20.160
So if that's what you're trying to do, sweet.
00:04:20.319 --> 00:04:22.160
Starts here with the live stream and the podcast.
00:04:22.240 --> 00:04:26.720
But if you haven't already, please make sure to go to our website at your everydayai.com.
00:04:26.800 --> 00:04:34.399
We're gonna be recapping not just the highlights from today's show, but everything else you need to know, especially the other AI news and developments, to stay ahead.
00:04:34.639 --> 00:04:38.240
And hey, actually a quick little announcement.
00:04:38.399 --> 00:04:40.319
It is back to school time.
00:04:40.399 --> 00:04:41.519
Yeah, here in the US.
00:04:41.600 --> 00:04:45.120
It's that time of the year where everyone goes back to school.
00:04:45.279 --> 00:04:51.040
But uh, maybe you're not a student or maybe you don't even have kids in school, but guess what?
00:04:51.199 --> 00:04:52.879
We're all going back to school together.
00:04:53.040 --> 00:04:57.600
Uh, so we're gonna be doing some learning together on everyday AI.
00:04:57.759 --> 00:05:07.439
So there's actually this thing that we've been doing for a while called our Start Here series that we kicked off earlier this year, and they've been some of our most popular shows ever.
00:05:07.680 --> 00:05:15.279
It covers all of the basics of not just you know yesterday's AI, but today's AI and where it's going.
00:05:15.360 --> 00:05:21.600
Uh, so we're gonna be running back the entire series starting this Monday with the first volume.
00:05:21.759 --> 00:05:36.079
So a little break from our normal Monday to Friday scheduling, but I guarantee if you stick with this series from volume one through volume 30 and you put the work in, you'll know more than about 99% of the other people in your company when it comes to AI.
00:05:36.160 --> 00:05:40.240
So no jargon, no fluff, no computer science degree required.
00:05:40.480 --> 00:05:43.199
Class starts Monday, don't be late.
00:05:43.360 --> 00:05:52.800
All right, let's get into today's episode, though, where we're gonna be talking about scheduling AI and how to easily make AI work for you, right?
00:05:52.879 --> 00:05:55.439
In Claude Gemini in Chat GPT.
00:05:55.519 --> 00:05:57.120
Yeah, because I don't know.
00:05:57.519 --> 00:06:00.319
I think there was a certain time before this was a feature.
00:06:00.480 --> 00:06:02.720
I was like, can we get this, please?
00:06:02.800 --> 00:06:03.040
Right.
00:06:03.120 --> 00:06:13.600
I remember seeing back in like uh 2024 when there was rumors, right, of these kind of scheduled tasks or you know, scheduled actions, they've gone under a lot of name changes.
00:06:13.759 --> 00:06:20.639
I remember when these were just leaks and rumors, and I'm like, oh my gosh, this will completely change the way the world uses large language models.
00:06:20.720 --> 00:06:21.680
And guess what?
00:06:21.920 --> 00:06:22.879
It didn't.
00:06:23.199 --> 00:06:33.439
Mainly because I think a lot of the companies kind of changed things around so much, or they buried these and made them hard to find, which is actually um interesting.
00:06:33.839 --> 00:06:50.160
Because I think that the uh kind of the agentic adoption was slower than it maybe could have been, if not if if these uh you know kind of scheduled options or features would have been earlier in front and center.
00:06:50.240 --> 00:06:55.199
Uh, because if I'm being honest, I was a very early adopter of these.
00:06:55.279 --> 00:06:56.480
Like the day it came out.
00:06:56.639 --> 00:06:59.680
I remember I think it was ChatGPT tasks that was the first to come out.
00:06:59.839 --> 00:07:01.839
It might have been in like late 2024.
00:07:02.000 --> 00:07:05.920
I started moving all of my workflows over to ChatGPT tasks.
00:07:06.160 --> 00:07:12.000
And I think that process, uh, you know, I always talk about this concept of unlearning and rebuilding.
00:07:12.160 --> 00:07:26.000
I think that was actually one of the most important steps for me to prepare myself for, you know, moving most of our processes and even my thinking, my creativity, my brainstorming into an agentic workflow.
00:07:26.079 --> 00:07:36.879
Because, you know, you really have to start deconstruct, uh, you know, to deconstruct your entire process when you're like, okay, wait, I can wake up and have all of this stuff done for me, right?
00:07:37.040 --> 00:07:42.240
You start, I think, uh thinking less in terms of what do I need to do next.
00:07:42.399 --> 00:07:45.199
Instead, you start thinking, what is the end goal?
00:07:45.279 --> 00:07:45.519
Right.
00:07:45.600 --> 00:07:51.519
And then you start working your way backwards from that and saying, how can I have this end goal waiting for me in the morning?
00:07:51.759 --> 00:07:52.319
All right.
00:07:52.800 --> 00:07:56.399
So uh let's go through the three main platforms.
00:07:56.560 --> 00:07:56.879
All right.
00:07:57.040 --> 00:08:03.680
And you know, if you don't know, the three main platforms, anyways, are uh Gemini, Claude, and Chat GPT.
00:08:03.759 --> 00:08:08.800
So yeah, technically, Copilot, it's just too messy to do this, right?
00:08:08.879 --> 00:08:16.319
I can show you the online version of Copilot, but almost every single one of our listeners, if you're using Copilot, you're using it on the desktop.
00:08:16.399 --> 00:08:31.439
The reason why I'm not doing Copilot, well, number one, I'm on a Mac, uh, but number two, uh, even if I was on a PC, it is entirely too difficult in terms of permissions, in terms of jumping through all of these different hoops.
00:08:31.600 --> 00:08:39.759
So it's almost like uh to show you these things, it would kind of be a waste because you would go out and be like, oh my gosh, this is amazing, and you would try and do it, and it wouldn't work.
00:08:39.919 --> 00:08:46.159
I actually did a show previously on the online version of Microsoft tasks, which I think are really good.
00:08:46.320 --> 00:08:49.279
So uh, you know, I'll make sure to drop that one in today's newsletter.
00:08:49.360 --> 00:08:51.840
But let's get into uh Google Gemini.
00:08:52.159 --> 00:08:59.840
So here's the quick rundown on uh Gemini's version, and these are called Gemini's Scheduled Actions.
00:09:00.080 --> 00:09:03.039
So let me just be honest, not that good.
00:09:03.200 --> 00:09:10.080
They're hard to find, they're buggy, it's extremely limited in terms of features and the user interface.
00:09:10.240 --> 00:09:13.440
And you also can't schedule via normal chat.
00:09:13.600 --> 00:09:18.320
And unfortunately, when you try to do that, you're actually gonna get some false positives.
00:09:18.639 --> 00:09:31.519
So the other downside is if you once you do get them scheduled, um, if you don't use them or check in on them, uh they will be automatically disabled, which stinks, right?
00:09:31.759 --> 00:09:33.840
Um, so yeah, a lot of downsides.
00:09:34.000 --> 00:09:36.159
So here's how you're going to find them.
00:09:36.240 --> 00:09:40.399
So if you just go to gemini.google.com, make sure you're logged in.
00:09:40.480 --> 00:09:43.600
So if you're looking on the left-hand panel, they're not gonna be there.
00:09:43.840 --> 00:09:44.159
All right.
00:09:44.320 --> 00:09:56.080
So, you know, fortunately, uh Chad GPT and Claude did uh move these uh scheduled tasks for them, that's what they're called, front and center, on the main nav bar.
00:09:56.240 --> 00:09:58.720
So uh Google Gemini, no such luck.
00:09:58.879 --> 00:10:03.519
So you have to actually look in your settings, and there you'll see these scheduled actions.
00:10:03.679 --> 00:10:04.159
All right.
00:10:04.480 --> 00:10:16.399
So once you do that, um, then essentially you are met with a very simple and straightforward enough um scheduled action box where you just essentially type it in.
00:10:16.559 --> 00:10:22.159
So you give it a name, you give it instructions, you say how often, and then a time.
00:10:22.399 --> 00:10:24.559
And that's it, right?
00:10:24.639 --> 00:10:35.440
It's an extremely simple interface, no bells, no whistles, and the downside is well, it's not that useful, if I'm being honest.
00:10:35.600 --> 00:10:36.000
All right.
00:10:36.159 --> 00:10:44.480
Uh, so next up, and I am gonna, I do have an example of you know, one kind of simple S uh prompt that I did.
00:10:44.639 --> 00:10:59.120
Um, and then at the very end, after I show you these examples, so yeah, this is one FYI and this is one is technically pre-recorded, uh, because it's gonna be it was a little difficult to, you know, go through uh all of my different systems without accidentally doxing myself.
00:10:59.279 --> 00:11:01.279
And also these are scheduled, right?
00:11:01.360 --> 00:11:07.519
So I didn't want to schedule these for, you know, oh, 745 and then oh, I'm done before then, right?
00:11:07.600 --> 00:11:10.639
Because you know, the live stream kicks off at 7:30 a.m.
00:11:10.799 --> 00:11:11.600
Central Standard time.
00:11:11.759 --> 00:11:16.000
Uh so it was gonna be a little tricky to try to get all these to schedule live and in real time.
00:11:16.159 --> 00:11:22.320
So yeah, this is one of those shows for AI at work on Wednesdays, normally very hands-on, doing live prompts, all that.
00:11:22.399 --> 00:11:28.159
So this is more of uh, you know, hey, I'm gonna show you how to cook, put it in the oven, and guess what?
00:11:28.240 --> 00:11:29.840
It's already done, got the cake ready.
00:11:30.000 --> 00:11:34.559
All right, so next up, let's talk about how ChatGPT's scheduling works.
00:11:34.799 --> 00:11:37.759
So theirs are called scheduled tasks.
00:11:37.919 --> 00:11:44.080
And again, I actually did a dedicated show uh on uh ChatGPT's scheduled tasks.
00:11:44.159 --> 00:11:51.600
Let me see what uh what one that was, because it was actually fairly recent because they went through some major changes.
00:11:51.759 --> 00:11:56.480
All right, so this was actually on July 1st, so episode 810.
00:11:56.639 --> 00:12:02.559
So I'm gonna go over the high levels, uh high level points for that, but it was actually a pretty in-depth episode.
00:12:02.639 --> 00:12:04.559
I remember I spent a good amount of time on it.
00:12:04.639 --> 00:12:10.799
Uh, so if you are a heavy chat GPT user, like a lot of people are, uh, make sure you go check out episode 810.
00:12:10.879 --> 00:12:12.159
But here's the gist.
00:12:12.399 --> 00:12:17.519
So tasks now support more agenc app actions, and that's the great thing.
00:12:17.679 --> 00:12:24.080
They also work inside of projects, and you can use memory for more personalized automation.
00:12:24.240 --> 00:12:25.519
That part's huge.
00:12:25.679 --> 00:12:38.240
Uh so you can access them directly from the sidebar, like I talked about before, or you can go to chatgbt.com/slash scheduled, and you can just create one by describing what you want ChatGPT to do in natural language.
00:12:38.480 --> 00:12:42.639
Like, you know, you can say scan my Gmail daily at 7 a.m.
00:12:42.799 --> 00:12:44.480
and tell me what's important.
00:12:45.759 --> 00:12:52.320
All right, but there's a lot more to uncover here because this is where ChatGPT really stands out.
00:12:52.480 --> 00:13:02.240
So a little um kind of hack, uh, you can go through the normal uh interface to set this up, or like I said, you can set it up conversationally.
00:13:02.320 --> 00:13:13.600
But what a lot of people don't understand is if you set this up conversationally, you actually have um some options that you don't have if you were to set it up in your normal scheduling routine.
00:13:13.759 --> 00:13:18.720
Uh, that's because, and I'm kind of looking on my on my other screen uh right now.
00:13:18.879 --> 00:13:32.559
So if you go into the scheduled section, um, you're gonna see it's it's gonna look like a fairly uh a fairly normal prompt box inside of Chat GPT, except there's one or two things missing.
00:13:32.639 --> 00:13:36.320
And the big thing here is the model selector.
00:13:36.480 --> 00:13:46.320
So yeah, it's um, I believe what happens in the schedule task, it just is gonna default to the last model you use, which may or may not be a good thing.
00:13:46.559 --> 00:14:00.559
So I recommend um just doing scheduling a task via natural language in the main kind of new chat box and just say schedule this at this time and chat GPT is smart enough where it's gonna do that.
00:14:00.720 --> 00:14:05.519
The main benefit there is well, you can choose the exact model you want.
00:14:05.679 --> 00:14:09.039
So maybe you want to use, you know, GPT-5.
00:14:10.240 --> 00:14:12.639
Soul extra high, right?
00:14:12.879 --> 00:14:17.919
At least you have the ability to configure the exact level of reasoning that you need for something.
00:14:18.000 --> 00:14:20.879
Uh, because maybe it's a little simpler, right?
00:14:21.039 --> 00:14:27.200
Maybe it's more of a personalized search and you don't exactly need the most, you know, the smartest model.
00:14:27.279 --> 00:14:37.200
But if I'm being honest, it's one of those things where uh unless you're running out of limits on a ChatGPT Pro plan, which is pretty much impossible.
00:14:37.360 --> 00:14:44.320
Um, it's you might as well just use the most powerful model at this because it's gonna be scheduled, it's gonna be happening while you sleep.
00:14:44.480 --> 00:14:46.879
So throw as much juice as you can at these things.
00:14:46.960 --> 00:14:51.200
Uh so that's a little tip there with the scheduling tasks inside ChatGPT.
00:14:51.360 --> 00:15:01.039
Uh, the other one, a little hack, and for whatever reason, this doesn't always work because maybe it is truly a hack.
00:15:01.279 --> 00:15:18.960
But if you go into the new work tab inside of Chat GPT, okay, uh, and I know that there's a lot of confusion about this, so I'm not gonna dive in too deeply, but essentially there's a new version of the Chat GPT desktop, which is called ChatGPT Work, which is kind of the same thing as codex.
00:15:19.200 --> 00:15:26.960
But then if you go to chat GPT.com, you'll see at the top now there's a new chat in a new uh work section.
00:15:27.120 --> 00:15:33.759
So the easiest way to tell the difference, because people are like, wait, when I go to chat gpt.com, do I use chat or do I use work?
00:15:34.000 --> 00:15:39.440
Well, the simplest way is well, use chat when you just want to chat or talk to Chat GPT.
00:15:39.600 --> 00:15:42.559
But if you want something more like an agent, use work.
00:15:43.120 --> 00:15:51.600
So the cool thing here, and this is again, I've never seen or heard anyone else doing this, and it doesn't work every single time.
00:15:51.679 --> 00:15:55.519
But if you get it to work once when scheduling your task, for me, it works over and over.
00:15:55.679 --> 00:16:04.080
But the cool thing in the work mode, it actually has its own virtual and dedicated computer and browser, which is awesome, right?
00:16:04.159 --> 00:16:11.759
So if there's anything that you normally have to do that requires a ton of interface clicking around, right?
00:16:11.919 --> 00:16:16.159
It could be from uh a website that you rely on every single day.
00:16:16.320 --> 00:16:22.159
It could just be a lot of you know web searching that you can't just do from reading a static page, right?
00:16:22.320 --> 00:16:32.080
Because a lot of times when these large language models, you know, quote unquote use the web, they essentially just scan the code um of any given page, which is helpful in most cases.
00:16:32.159 --> 00:16:35.679
But what if you have to, you know, scroll down to load more?
00:16:35.840 --> 00:16:41.360
Uh, what if you have to um you know click around a bunch and add filters, right?
00:16:41.440 --> 00:16:47.440
So think of all the web browsing that you do that maybe you haven't yet handed off um to an agent.
00:16:47.519 --> 00:16:49.440
And there's probably a reason for that.
00:16:49.679 --> 00:16:51.440
And this is awesome.
00:16:51.600 --> 00:16:54.480
So, you know, one quick example of something like this.
00:16:54.960 --> 00:16:57.039
I love Google Trends, right?
00:16:57.120 --> 00:17:01.840
So there's something in Google Trends, you go, you can put all these filters on, and then you have to scroll down.
00:17:01.919 --> 00:17:05.519
I like to see like what's trending in big tech in AI.
00:17:05.759 --> 00:17:08.960
And I realized, well, why am I doing that?
00:17:09.119 --> 00:17:09.359
Right.
00:17:09.440 --> 00:17:40.319
So that's something I went into Chat GPT, clicked the work, uh, did just a natural language, gave it the URL to that, you know, once I put all those filters on uh the Google trend, and then I said, Hey, every day at 9 p.m., go to this URL, scroll down to the bottom, uh, see, you know, use everything that you know about me and the everyday AI audience and see which of these trends that our audience would care the most about, then go out and research all those trends, tell me why they're trending, right?
00:17:40.480 --> 00:17:45.359
Like uh yesterday something about the iPhone was trending and I didn't know why, right?
00:17:45.440 --> 00:17:55.279
So it's gonna go and it's gonna, you know, help me stay more in touch with things that matter to me in my career without me having to go through and click and then go out and research.
00:17:55.359 --> 00:17:57.519
I just wake up and I know, right?
00:17:57.599 --> 00:18:03.279
So that's actually something inside of Chad GPT that is very unique to it because uh, you know, Claude can't do that.
00:18:03.440 --> 00:18:06.720
Uh, you know, we're not talking co-pilot, co-pilot can't do that.
00:18:06.799 --> 00:18:09.759
And obviously, Gemini's scheduled actions can't do that.
00:18:09.839 --> 00:18:10.960
So big hack there.
00:18:11.119 --> 00:18:16.079
Um, another one that's very unique to Chad GPT is having these open-ended tasks.
00:18:16.160 --> 00:18:19.759
So this is something that uh OpenAI didn't even really highlight.
00:18:19.839 --> 00:18:22.960
Um, in I think it was technically late June that this came out.
00:18:23.039 --> 00:18:26.160
We covered it like the next day, I think it was like July 1st, right?
00:18:26.559 --> 00:18:32.799
Um, but you can now just say something like, you know, check my competitor's pricing page and tell me when it changes.
00:18:32.960 --> 00:18:35.920
You don't have to say check it every Tuesday or at this PM.
00:18:36.400 --> 00:18:40.640
I don't know how this magic works, but it just does, right?
00:18:40.799 --> 00:18:42.000
Uh so that's really cool.
00:18:42.079 --> 00:18:44.559
You can also just give it open-ended tasks.
00:18:45.359 --> 00:18:46.000
All right.
00:18:46.400 --> 00:18:49.200
Um, so now I'm showing on my screen how that works.
00:18:49.279 --> 00:18:56.480
So on the left-hand side, you have your um uh scheduled tasks there.
00:18:56.640 --> 00:19:07.039
Uh, then uh you can go in to the scheduled task, and like I said, you still get uh you know, you can use all your connectors, your skills, all of those drop down.
00:19:07.119 --> 00:19:20.240
But the downside is you don't have that model selector, but you can still create a scheduled task from the scheduled section, or you can just click new chat and just type it out um in natural language, give a time, a frequency, etc.
00:19:20.799 --> 00:19:23.440
Um, then you can go in and edit them.
00:19:23.519 --> 00:19:32.240
So if you when you do click that uh scheduled portion, or if you go to chat gpt.com slash scheduled, uh, you can go in and edit it.
00:19:32.319 --> 00:19:39.279
So if you do give a prompt in natural language, it's gonna turn that into a set of better instructions for the model to follow.
00:19:39.440 --> 00:19:42.000
But then from there, you can make any edits.
00:19:42.079 --> 00:19:50.880
Uh so if you remember the kind of old school uh GPT builder, you could just chat with ChatGPT and it would turn your chats into a set of custom instructions.
00:19:51.039 --> 00:19:53.279
Kind of the same thing happens here under the hood.
00:19:53.359 --> 00:19:57.839
So you can go in and update the end uh scheduled task.
00:19:57.920 --> 00:20:05.839
And you can also, there's a drop-down menu for when you want it to repeat, the frequency, the time, and an end date.
00:20:06.799 --> 00:20:09.839
All right, let's keep going and talk about clauds.
00:20:10.000 --> 00:20:14.480
So clauds are powerful, but a little confusing.
00:20:14.640 --> 00:20:16.240
Okay, and here's why.
00:20:16.400 --> 00:20:19.519
Uh, because you have your normal scheduled tasks.
00:20:19.599 --> 00:20:22.480
So again, this is all just Claude on the web.
00:20:22.720 --> 00:20:25.680
Okay, so I'm not talking, you know, codex desktop.
00:20:25.759 --> 00:20:30.000
I'm not talking about, you know, Claude Desktop with Claude Code and Claude Cowork.
00:20:30.079 --> 00:20:31.839
Uh, I am talking about Claude on the web.
00:20:32.079 --> 00:20:39.279
But if you didn't know this, Claude on the web also changed a little bit, just kind of like how ChatGPT did, how you have chat and work.
00:20:39.359 --> 00:20:42.799
Uh, so now Claude has home in code.
00:20:42.960 --> 00:20:48.400
All right, so if you go to home, uh you can do scheduled tasks that way.
00:20:48.480 --> 00:20:55.440
So you essentially can do a scheduled task in normal chat or in Claude cohort, even more confusing.
00:20:55.519 --> 00:20:58.319
So there's technically like three different ways you can set this up.
00:20:58.480 --> 00:21:02.880
Uh, but uh the other way is a routine.
00:21:03.039 --> 00:21:07.839
So if you click on the Claude Code button, you can set up a routine.
00:21:07.920 --> 00:21:11.680
And I'm gonna quickly tell you uh the difference here in the pros and the cons.
00:21:11.839 --> 00:21:22.480
For for Claude's scheduled tasks, um, that's just the uh it is very much the equivalent of the ChatGPT version, which is like chat and work, right?
00:21:22.559 --> 00:21:27.440
So if you need something that's just more straightforward, you know, just use the the home.
00:21:27.519 --> 00:21:32.400
Uh, if you need something that's a little more technically uh agentic, you can click the co-work tab.
00:21:32.559 --> 00:21:41.200
But you can also, just like you can in ChatGPT, you can create a scheduled task just conversationally with Claude, or there is a manual uh setup.
00:21:41.359 --> 00:21:44.480
Uh, the cool thing about Clauds is you can work in a project.
00:21:44.640 --> 00:21:56.640
You can also set an approval level per task, because um, as an example, all right, if if you're giving it send actions for the first time it runs, you might want to approve those.
00:21:56.720 --> 00:21:59.759
Uh, I don't fully understand uh if you're trying.
00:22:00.160 --> 00:22:01.200
To schedule, right?
00:22:01.279 --> 00:22:09.519
I kind of scratched my head on this because I'm like, okay, why would you actually want uh you know to like manually approve everything if it's on a schedule, right?
00:22:09.599 --> 00:22:14.640
Like if you have 10 of these things running overnight, why would you not rent them auto-approved?
00:22:14.720 --> 00:22:17.440
So I guess in theory, it's a nice feature.
00:22:17.680 --> 00:22:20.240
I just literally don't know the point of it.
00:22:20.400 --> 00:22:21.920
But, anyways, it's there.
00:22:22.079 --> 00:22:27.359
Maybe if you just want to run it once in front of you to see how it works, maybe that's what it's there for.
00:22:27.519 --> 00:22:36.559
Um, otherwise, it's extremely counterintuitive to schedule something, but then oh, it's not gonna fully run because you had to approve it.
00:22:36.799 --> 00:22:42.480
Anyways, uh, you can also choose the model, but you cannot unfortunately choose the reasoning level.
00:22:42.640 --> 00:22:55.839
You can set the frequency, it doesn't have a custom uh version like ChatGPT, but you can also uh use connectors and skills like you can in Chat GPT, but then you have these scheduled routines.
00:22:55.920 --> 00:22:58.160
And I actually did this, let me see.
00:22:58.240 --> 00:23:02.319
This this show was a little uh a little longer ago.
00:23:02.480 --> 00:23:03.680
Let me see when I did this.
00:23:03.839 --> 00:23:06.319
The scheduled uh routines.
00:23:06.480 --> 00:23:10.880
Yeah, this is live all or you know, unedited, unscripted at uh at least.
00:23:11.119 --> 00:23:12.880
So let's see when I did this show.
00:23:13.039 --> 00:23:20.079
Um so this was okay, this was episode 756.
00:23:20.640 --> 00:23:34.319
Um this was in April, but these are all uh still very, very relevant because um, you know, we went over this was technically through the scope of Claude Desktop, but it kind of worked the exact same.
00:23:34.480 --> 00:23:41.839
But we talked in depth about the differences uh between these scheduled tasks and the routines in Claude Code.
00:23:41.920 --> 00:23:48.400
But the difference on the routine, um, right, is well, you can actually sync those with your desktop.
00:23:48.480 --> 00:23:56.960
Uh so if you set up a Claude Schedule task, it's not gonna sync inside Claude um desktop.
00:23:57.039 --> 00:24:03.279
So the only way you can do it is with a Claude Code routine, and then you uh have to choose the cloud version.
00:24:03.440 --> 00:24:11.839
So um on codex, uh right on the desktop version, uh, if you're using ChatGPT work to schedule, it'll all sync.
00:24:12.079 --> 00:24:15.519
Uh so why would you use Claude's scheduled routines?
00:24:15.599 --> 00:24:20.079
And I'll just give you the 30-second version of uh episode 756.
00:24:20.240 --> 00:24:21.519
I think they're really, really good.
00:24:21.680 --> 00:24:31.519
Um, they're more powerful, and I think a lot of people, especially non-technical people, are gonna look at like Claude routines and be like, uh, Claude code and be like, I'm not touching those, right?
00:24:32.400 --> 00:24:37.279
You're gonna say there's enough with skills and all these other things and and co-work, right?
00:24:37.359 --> 00:24:38.720
That's technical enough for me.
00:24:38.960 --> 00:24:41.759
The claw claude routines are super simple.
00:24:41.920 --> 00:24:47.599
The you know, you you don't even really have to use clawed code to go in and use claud routines.
00:24:47.839 --> 00:24:57.599
Um, the the the biggest um kind of benefit is you can use um essentially an API token as a trigger.
00:24:57.759 --> 00:25:00.160
Uh, so what that means, it's very simple.
00:25:00.240 --> 00:25:05.680
Uh, you can choose that, and I have a screenshot of that later so I can uh walk through.
00:25:06.079 --> 00:25:07.359
Oh, I have it right here.
00:25:07.519 --> 00:25:14.000
Um, so essentially, if you choose that as an option, you're gonna get essentially an API token.
00:25:14.160 --> 00:25:16.640
All right, and then you can go to something like Zapier.
00:25:16.880 --> 00:25:27.759
Um, and you can use that to essentially connect anything in Zapier and have that be a trigger to start something in a Claude routine, which is really cool, right?
00:25:27.839 --> 00:25:42.960
Because as an example, you know, in ChatGPT, they actually used to have this, I believe, in Codecs, but then they kind of simplified and unified uh the the codecs in Chat GPT uh they used to be called automations, so now they're just called schedule tasks.
00:25:43.119 --> 00:25:45.119
Anyways, this used to be available in codecs.
00:25:45.200 --> 00:25:49.200
I don't believe it is anymore, but this is great in uh claude routines.
00:25:49.440 --> 00:26:00.480
So, as an example, you can set anything uh that happens in Zapier as a trigger to start um to start a routine inside of Claude code.
00:26:00.640 --> 00:26:02.480
Again, you don't have to be super technical.
00:26:02.640 --> 00:26:05.839
You can literally just give it to Claude and be like, walk me through this.
00:26:05.920 --> 00:26:13.599
Um, right, but uh that's super valuable because otherwise, every other thing has to start at a certain time, right?
00:26:13.759 --> 00:26:23.839
So maybe I don't know, you you have a super uh you know, um highly valuable uh form on your company's website, right?
00:26:23.920 --> 00:26:30.720
Uh when someone comes in, it's a potential million-dollar client, and you want to be able to start doing some legwork right away when it comes in.
00:26:30.880 --> 00:26:34.720
Obviously, you can't do a schedule task on that because you have no clue.
00:26:34.880 --> 00:26:42.960
But in this case, right, you can set this up inside Claude Code, takes like 30 seconds, choose that as an option, you're gonna get this key.
00:26:43.200 --> 00:26:49.039
Then you go into Zapier, you know, whatever it is for using a Google form, Jot form, whatever.
00:26:49.119 --> 00:26:52.000
Uh, you know, then you're gonna get this key, plug it into Zapier.
00:26:52.160 --> 00:27:00.400
And then anytime that web form comes in, then that routine is going to start inside of Claude Code once it kind of sends to that API token.
00:27:00.559 --> 00:27:06.000
So uh a nice unique feature there inside the uh scheduled routines.
00:27:06.559 --> 00:27:09.519
All right, so uh that's kind of the gist of it.
00:27:09.680 --> 00:27:15.839
And I'm gonna go through now and uh I'm gonna share my screen for our live stream audience and podcast audience.
00:27:15.920 --> 00:27:20.240
You can always go to our website at your everydayai.com if you want to see the video version of this.
00:27:20.400 --> 00:27:23.920
Nothing super visual, and I'll kind of explain what's going on.
00:27:24.079 --> 00:27:24.880
Um, right.
00:27:24.960 --> 00:27:31.440
And or you can always check the show notes and we uh do uh put a link directly to today's episode.
00:27:31.599 --> 00:27:35.759
All right, so uh let's talk about what our test actually was.
00:27:35.920 --> 00:27:38.240
So this one was simple-ish.
00:27:39.039 --> 00:27:39.839
Um, right.
00:27:40.319 --> 00:27:50.799
The downside is because Google Gemini uh is not good at even using its own connectors like Gmail.
00:27:50.880 --> 00:28:01.039
Um, I was trying to find something that I thought in theory uh could be a very um, you know, apples to apples comparison and something maybe realistic.
00:28:01.200 --> 00:28:04.480
And also this is something that I wanted to build, anyways.
00:28:04.640 --> 00:28:10.480
Um, so I said, all right, let's just throw a simple schedule prompt um and talk about the use case of this.
00:28:10.720 --> 00:28:15.039
So um, you know, I said every day at blank o'clock, right?
00:28:15.119 --> 00:28:18.799
Uh, because I spaced these out a little bit, but I gave it the same thing.
00:28:18.880 --> 00:28:28.079
Uh so I said every day at this time, please put together an interactive either site or artifact or canvas, uh, because they all call them something different.
00:28:28.160 --> 00:28:32.799
But this is essentially an interactive visual, and in some it worked and some it didn't, and I'll show you that.
00:28:32.960 --> 00:28:52.319
Um, so at every day at uh a certain time, please put together an interactive visual that shows the current stock prices for these AI stocks NVIDIA, Alphabet, Microsoft, Broadcom, Meta Platforms, uh, Taiwan Semiconductor, Micron Technologies, uh, AMD, Oracle, Palantir.
00:28:52.640 --> 00:29:01.039
Uh, also, please include an up-to-date factual bullet point recap for each that explains the biggest news for each of these companies.
00:29:01.279 --> 00:29:13.759
So, this is something that I don't do it every single day, but I probably every single day end up checking at least two to three um of these companies in terms of what's happening with their stock.
00:29:14.000 --> 00:29:27.359
Not because I, you know, have stock in all of them and I'm tracking it from a portfolio perspective, but from an AI news perspective, a lot of times a company stock will shoot up five percent or drop five percent, which is very rare.
00:29:27.440 --> 00:29:31.759
And a lot of times it has to do with something that's happening in AI, as an example.
00:29:31.920 --> 00:29:50.480
You know, Alphabet slash, you know, the parent company of Google uh announced a major uh AI leadership change um, you know, a couple of days ago, and their stock dropped like five percent, which for a multi-trillion dollar market cap company is an insane amount of money, right?
00:29:50.559 --> 00:29:55.519
So sometimes I just want to be able to see at a glance what's happening with all these big AI companies.
00:29:55.599 --> 00:29:59.839
And you know, essentially if I see a 3% up or down, that's kind of my threshold.
00:30:00.160 --> 00:30:05.359
Then I know that something meaningful is happening probably in AI around news.
00:30:05.519 --> 00:30:07.839
So this is something I kind of wanted to build for myself.
00:30:08.000 --> 00:30:10.960
But the thing is, well, it refreshes all the time.
00:30:11.039 --> 00:30:20.640
So that's the good thing with this being a scheduled right, this is now technically an agent that's going out and doing this every single day for me and updating it so I don't have to.
00:30:20.880 --> 00:30:26.480
All right, so let's go ahead and look at the results.
00:30:27.119 --> 00:30:28.960
Let's share my screen here.
00:30:29.200 --> 00:30:34.799
All right, so Google Gemini first and Google Gemini kind of failed.
00:30:34.960 --> 00:30:36.960
All right, so what's interesting?
00:30:37.119 --> 00:30:42.480
So I did say use canvas mode, um, but it didn't.
00:30:42.799 --> 00:31:03.200
So um I do believe I could be mistaken, but I do believe when you used to have uh the um scheduled actions um here inside of uh Google Gemini, I thought you used to be able to use some of the other features, specifically uh the canvas mode.
00:31:03.599 --> 00:31:06.319
Uh, but in this case, it didn't work.
00:31:06.559 --> 00:31:08.720
Um, so that's interesting.
00:31:08.880 --> 00:31:11.519
It did, however, go out and do the rest.
00:31:11.680 --> 00:31:14.240
So it actually fired at the correct time.
00:31:14.400 --> 00:31:16.799
So I did this one at 5 15 p.m., right?
00:31:16.880 --> 00:31:18.559
So after the markets closed.
00:31:18.720 --> 00:31:24.880
Um, so it got the correct prices for NVIDIA, and then it gave me the key news, alphabet, Microsoft, etc.
00:31:25.279 --> 00:31:28.079
So it did fire at the correct time.
00:31:28.240 --> 00:31:39.119
It followed instructions, um, it uh, you know, got the correct prices, but it didn't put together a visual uh that I wanted to use in Canvas mode, right?
00:31:39.200 --> 00:31:50.160
I could go down here um and I could say, right, I'll actually do this just so we can actually just make it at least give Google and Gemini a chance because otherwise I'm just ragging on Gemini this entire time.
00:31:50.319 --> 00:31:53.839
So I'm gonna say, uh, you know, using all this info.
00:31:54.799 --> 00:31:55.119
All right.
00:31:55.279 --> 00:32:00.480
So yeah, unfortunately, I'm gonna have to um put together a two-step one on this.
00:32:00.640 --> 00:32:09.839
So uh this is where I also wanted to, you know, test out and say, hey, what other features um are available when you're scheduling a task?
00:32:10.000 --> 00:32:15.200
Because sometimes uh uh features or modes make all the difference.
00:32:15.359 --> 00:32:20.079
It's the difference between, you know, using your data and not using your data, uh, right?
00:32:20.160 --> 00:32:30.160
Like spoiler alert, one of the reasons I'll I've never really used uh Google Google Gemini's uh scheduled tasks is because they don't use your connected data, right?
00:32:30.240 --> 00:32:32.720
So it's like for me, it's not super useful.
00:32:32.880 --> 00:32:37.599
And in this case, it wasn't able to use the mode um for Canvas.
00:32:37.759 --> 00:32:41.359
All right, next, let's go ahead and take a look.
00:32:42.400 --> 00:32:44.319
Uh let's take a look at Claude.
00:32:44.880 --> 00:32:46.559
So uh same thing.
00:32:46.720 --> 00:32:51.119
Um, I did tell it to use an artifact.
00:32:51.359 --> 00:32:53.920
It however didn't.
00:32:54.079 --> 00:32:59.839
So it just wrote some strict HTML, but I don't know.
00:32:59.920 --> 00:33:03.839
Um, I'm not gonna say it's a failure because it looks good enough, uh, right?
00:33:04.000 --> 00:33:10.240
It actually put together a uh very clean uh, you know, it says AI stocks daily.
00:33:10.319 --> 00:33:11.680
It's interactive, right?
00:33:11.759 --> 00:33:14.319
It has a light mode, a dark mode.
00:33:14.400 --> 00:33:20.720
Uh, it gave me very clear ticker symbols, uh, the percentage that they're up and down on the day, right?
00:33:20.799 --> 00:33:22.799
Like here I see Google is down 3%.
00:33:23.440 --> 00:33:26.720
So I can click there and see what happened, right?
00:33:26.799 --> 00:33:34.400
So it said, you know, Tuesday it fell uh, you know, 3.8% amid a rotation of high valuation mega cap tech.
00:33:34.720 --> 00:33:38.799
It says investors remain uneasy about AI spending, right?
00:33:38.880 --> 00:33:45.440
As their cap X guidance was uh about 200 billion, uh, and their capex was up 100% year over year.
00:33:45.680 --> 00:33:49.440
So this for me, this is all really helpful stuff to read, right?
00:33:49.680 --> 00:33:54.000
So when I'm talking about AI things, I you know can ground it in a little more recent data.
00:33:54.160 --> 00:33:59.920
Uh, there's also this button that it says, you know, uh expand all news, collapse all news, which is really nice.
00:34:00.079 --> 00:34:04.240
So it didn't technically make it like a shareable artifact.
00:34:04.400 --> 00:34:09.199
Um, but overall, you know, it it did it did pretty good.
00:34:09.360 --> 00:34:15.519
Um so uh Gemini, you know, probably failure.
00:34:15.679 --> 00:34:20.480
Uh you know, uh Claude got a passing grade, but not ideal.
00:34:20.719 --> 00:34:23.599
But let's take a look at Chat GPT.
00:34:23.840 --> 00:34:26.639
So my gosh, this one's good.
00:34:26.719 --> 00:34:27.039
All right.
00:34:27.119 --> 00:34:36.960
So yeah, if if if you're not looking um at the live stream here, this is one of those ones where you know, sometimes I'm like, oh yeah, it's cool.
00:34:37.039 --> 00:34:38.559
This one's really good.
00:34:38.800 --> 00:34:39.920
This one's really good.
00:34:40.079 --> 00:34:43.199
All right, for a zero shot prompt, right?
00:34:43.280 --> 00:34:44.719
I read it to you, super simple.
00:34:44.960 --> 00:34:47.280
Um did a fantastic job.
00:34:47.360 --> 00:34:51.519
So I used ChatGPT sites, which by the way is a freaking cheat code.
00:34:51.599 --> 00:34:55.039
Like, if you're not using ChatGPT sites, it is in ChatGPT now.
00:34:55.119 --> 00:34:59.679
It used to only be available um inside of uh codecs.
00:34:59.840 --> 00:35:02.880
It used to only be available for team plans.
00:35:03.039 --> 00:35:07.599
Now it's available even on the web and it's available on individual paid plans as well.
00:35:07.760 --> 00:35:12.079
If you're not using ChatGPT sites, I'm just saying this, you're literally missing out.
00:35:12.400 --> 00:35:14.639
I make them all the time, right?
00:35:14.719 --> 00:35:20.480
I have custom skills uh to make a lot of them in kind of a templated way that really work for me.
00:35:20.639 --> 00:35:22.800
Uh, I send them to my wife, right?
00:35:22.880 --> 00:35:24.079
Because you can make them public.
00:35:24.239 --> 00:35:28.800
Uh, anyways, let's take a look at what uh Chat GPT sites did.
00:35:29.360 --> 00:35:33.760
Uh first of all, visually, fantastic, right?
00:35:33.920 --> 00:35:35.119
It gave me this breakdown.
00:35:35.360 --> 00:35:38.480
You know, it says four of the 10 advanced, six of the 10 declined.
00:35:38.559 --> 00:35:41.920
It gave me what's called an AI basket pulse, which is pretty cool.
00:35:42.079 --> 00:35:45.920
So it said overall, it was down 0.6% across 10 companies.
00:35:46.000 --> 00:35:49.360
So I can see overall, you know, how are these things moving?
00:35:49.679 --> 00:35:52.000
Give me the winners, the losers, uh, right?
00:35:52.079 --> 00:35:54.639
I can click on the biggest moves, the winners first.
00:35:54.719 --> 00:35:57.840
Uh, I can create a watch list, uh, which is pretty cool, right?
00:35:57.920 --> 00:36:00.239
I can toggle this little star on and off.
00:36:00.320 --> 00:36:09.119
Um, and then when I click on any of these, um it then brings the uh all of the news that mattered.
00:36:09.199 --> 00:36:10.079
So let's find another.
00:36:10.159 --> 00:36:11.440
We already talked about Google.
00:36:11.679 --> 00:36:17.280
Or no, actually, let's let's click Google and see if it gave me similar uh reasons that Google was down almost 4%.
00:36:17.679 --> 00:36:23.199
So it says Alphabet reported Q2 revenue of 120 billion, up 24 year over year.
00:36:23.280 --> 00:36:36.079
Search search and other revenue rose 17, uh Google Cloud revenue grew 82%, and backlog reached 514 billion with management attributing the acceleration to AI infrastructure in AI solutions.
00:36:36.159 --> 00:36:38.000
Uh, so I will say this.
00:36:38.320 --> 00:36:41.519
Uh, it pulled some older news, right?
00:36:41.679 --> 00:36:48.320
So uh if I were to redo this, uh obviously the ChatGPT one did the best, uh, but it pulled in some older news.
00:36:48.480 --> 00:36:57.920
Uh, I don't know if the Google one didn't give me the date of the news, um, but I'm guessing this news was maybe a little more recent.
00:36:58.079 --> 00:36:58.960
I'm not sure.
00:36:59.119 --> 00:37:05.679
Uh, but from a visual uh UI UX perspective, it's just much better, right?
00:37:05.840 --> 00:37:13.679
I'm actually gonna go through and uh, you know, probably make one or two changes in this and you know, make it mobile optimized.
00:37:13.760 --> 00:37:18.880
And the good thing is, is then all I do, you know, on my phone, I can essentially save this as an app.
00:37:18.960 --> 00:37:26.320
And then, you know, I can save it to my home screen, you know, and every day when I open it, it's gonna reload and bring up the freshest information.
00:37:26.559 --> 00:37:36.159
So um yeah, overall, uh I think they all had their own strengths, right?
00:37:36.239 --> 00:37:37.360
Uh, let's take a look.
00:37:37.440 --> 00:37:38.719
Um, going back over.
00:37:39.840 --> 00:37:43.199
Uh, sure enough, uh, we had another error.
00:37:43.360 --> 00:37:46.320
Um, another error in uh Google Gemini.
00:37:46.480 --> 00:37:46.800
That's fine.
00:37:46.880 --> 00:37:51.760
I really try to give uh Gemini in Canvas mode, uh, which I've always loved their canvas mode.
00:37:51.840 --> 00:37:52.480
I think it's really good.
00:37:52.559 --> 00:37:53.119
There we go.
00:37:53.199 --> 00:37:54.960
Uh so now it's working.
00:37:56.079 --> 00:37:57.039
Is it okay?
00:37:57.199 --> 00:37:57.679
Interesting.
00:37:57.840 --> 00:37:59.519
So it went a completely different direction.
00:37:59.679 --> 00:38:04.079
Give me kind of bar charts, uh, you know, prices, quarter two growth, capex.
00:38:04.480 --> 00:38:06.960
Uh so you know, it actually looks nice.
00:38:07.039 --> 00:38:11.920
Uh, some nice features and functionalities, not exactly what I was expecting.
00:38:12.000 --> 00:38:16.079
Uh, but then okay, when I scroll down, uh, I think it does a pretty good job.
00:38:16.159 --> 00:38:17.440
Let's read Google.
00:38:17.679 --> 00:38:22.480
Um, so it says uh executive recap summary.
00:38:22.719 --> 00:38:26.639
Although the Google it's not right, right?
00:38:26.719 --> 00:38:34.239
Uh, because it says it's it's up one percent, uh almost one percent for the day, which um I don't believe it was right now.
00:38:34.400 --> 00:38:37.280
I now I have to double check uh to make sure.
00:38:37.440 --> 00:38:39.440
Yeah, Google was down almost four percent.
00:38:39.599 --> 00:38:46.960
So Google was wrong about the actual uh day, so that kind of ruins it, right?
00:38:47.039 --> 00:38:53.760
So I'm not even gonna look at the rest because if you get the facts wrong, right, you always gotta you always gotta check these things, right?
00:38:53.840 --> 00:39:03.840
But yeah, uh Google was actually down you know, three point, you know, roughly seven percent, which the other uh models got correct, and Google unfortunately got it wrong.
00:39:03.920 --> 00:39:10.079
So even though you know it did finish the dashboard with multiple prompts, get a fail, Google.
00:39:10.159 --> 00:39:10.880
We get a fail.
00:39:10.960 --> 00:39:11.360
All right.
00:39:11.599 --> 00:39:15.440
So let's wrap up here and talk about my takeaways.
00:39:15.599 --> 00:39:18.239
So overall, Gemini, not useful at all.
00:39:18.400 --> 00:39:20.400
It struggles using its own connectors.
00:39:20.559 --> 00:39:27.599
Uh essentially, the only thing that it's really useful for is to customize web search for you.
00:39:27.760 --> 00:39:33.440
Uh, but even like you saw here, when in my again, my one example, it pulled in incorrect information.
00:39:33.679 --> 00:39:36.719
So uh I haven't used Google's, right?
00:39:36.800 --> 00:39:42.559
Aside from when I test these things, I haven't used Google's um scheduled actions.
00:39:42.639 --> 00:39:43.920
I hope they improve them.
00:39:44.000 --> 00:39:53.440
Uh, right, we'll see uh when uh we get a whether it's a Gemini 3.7 Pro, Gemini 3.5 Pro, Gemini 4 Pro, whatever.
00:39:53.519 --> 00:39:56.719
Uh hopefully they bring some updates to the Gemini web platform.
00:39:56.800 --> 00:39:58.639
They did go through a UI UX update.
00:39:58.719 --> 00:40:02.480
So hopefully these scheduled actions uh get some love as well.
00:40:02.559 --> 00:40:14.400
But overall, I'm not using them and I would say you probably shouldn't either, unless Gemini is literally the only thing that you can use at work, and the only thing you really need is like a customized Google search.
00:40:14.480 --> 00:40:16.719
Um, because otherwise, really not useful.
00:40:16.800 --> 00:40:17.280
Uh, all right.
00:40:17.360 --> 00:40:20.719
My main Chad GPT takeaway is it's really good.
00:40:20.880 --> 00:40:25.519
There's a lot of unique benefits, like the you know, the any-time scheduling, right?
00:40:25.599 --> 00:40:30.880
Uh just hey, go check when this happens and it somehow figures it out without you having to schedule it.
00:40:31.119 --> 00:40:35.679
Um, using work mode with the virtual computer and browser is really good.
00:40:36.000 --> 00:40:40.559
Um, the custom schedule the others don't have, like if you say like every other Tuesday, right?
00:40:40.639 --> 00:40:44.000
You can't do that um in Claude or in Gemini.
00:40:44.320 --> 00:40:54.719
Uh having an end date is a unique feature as well in ChatGPT because maybe as an example, you want to run it every single day for two weeks, but then you don't want it to keep running and clog up your feet.
00:40:54.880 --> 00:40:55.679
Uh, it's great.
00:40:55.840 --> 00:41:07.599
I think ChatGPT, uh, their schedule tasks more than any others are great at contextualizing um your history or contextualizing your schedule task in your data and history, right?
00:41:07.760 --> 00:41:11.760
I forgot the technical uh update for ChatGPT's memory.
00:41:11.840 --> 00:41:14.559
I think it's called like Dream V3 or something.
00:41:14.800 --> 00:41:20.079
It is so much better, so much better than you know anyone else out there.
00:41:20.159 --> 00:41:21.519
It's not even close.
00:41:21.679 --> 00:41:24.800
Um, I again don't know how OpenAI does it.
00:41:24.960 --> 00:41:28.079
Um, but that actually really comes into play.
00:41:28.159 --> 00:41:30.400
Um, when you're scheduling these things, right?
00:41:30.559 --> 00:41:38.719
Especially when they're of high value, you want to be able to bring in all of that other context and history from everything that ChatGPT or Claude knows about you.
00:41:38.880 --> 00:41:50.079
So unfortunately, right, like if you're doing the Claude routine, uh, which is really helpful, Claude Routine has no clue what you've been doing in Claude Chat or in Claude Cowork.
00:41:50.239 --> 00:41:54.159
So you can't wrap it in that extremely important business context.
00:41:54.320 --> 00:41:59.360
All right, ChatGPT, you can also use your connectors and plugins and skills, which is really cool.
00:41:59.599 --> 00:42:04.239
Um then, like I said, being able to tap into the full power of ChatGPT.
00:42:04.320 --> 00:42:10.719
So both in the reasoning levels is extremely important and helpful, or being able to use the sites feature, right?
00:42:10.800 --> 00:42:13.599
Having that sites feature that just updates all the time, amazing.
00:42:13.760 --> 00:42:19.280
All right, so now let's talk about Claude, the home or the chat scheduling takeaway.
00:42:19.440 --> 00:42:22.559
So more straightforward, but it works, right?
00:42:22.719 --> 00:42:26.000
Unlike Google, you can use all your connectors, which is good.
00:42:26.159 --> 00:42:29.119
Being able to set the permissions, even though I don't fully understand that.
00:42:29.199 --> 00:42:32.079
If you want to run it once as a test, maybe that's helpful.
00:42:32.320 --> 00:42:45.039
Overall, straightforward, nothing, you know, there's no standout features in the um kind of home or co-work scheduling, but there is um in the other version in the routines.
00:42:45.119 --> 00:42:51.360
So in the routines, there's a lot of, you know, I already went over the um kind of the two, well, two unique features.
00:42:51.440 --> 00:42:52.400
I didn't mention the other one.
00:42:52.480 --> 00:43:04.239
The one is being able to kind of have that API token um and have that act as a trigger, which is very unique in terms of what you can do to schedule right now inside of Claude Code routines.
00:43:04.400 --> 00:43:06.719
Like I said, don't be scared of Claude Code.
00:43:06.800 --> 00:43:10.239
You don't even have to use anything else in Claude Code, uh, right?
00:43:10.400 --> 00:43:17.599
You can just think of it like a normal chat, but you can kick off um a routine based on literally anything that happens on the web.
00:43:17.679 --> 00:43:29.920
Just go in Zapier, you know, take that API token and say anything that happens in the 6,000 platforms that Zapier um connects with, send it to this API token, and then you can kick off any routine.
00:43:30.079 --> 00:43:36.239
Uh, a great um a great use case and something unique for all of the others when it comes to scheduling your work.
00:43:36.400 --> 00:43:48.639
And then the other one, which is a small one, um, inside clawed um claud code routines, you can also set routine level messaging alerts, which sounds like okay, why does it matter?
00:43:48.800 --> 00:43:55.599
Um, you know, in ChatGPT and uh Gemini, you can set notifications for all of them, but not at the individual level.
00:43:55.679 --> 00:44:05.119
So one of the reasons why that's helpful, uh, right, is you can, as an example for a certain routine that runs, you can send a Slack message, uh, you know, which is pretty helpful.
00:44:05.440 --> 00:44:05.840
All right.
00:44:06.000 --> 00:44:08.719
So, like I said, there's pros and there's cons.
00:44:08.880 --> 00:44:16.559
I think overall, um, Chat GPT is the winner with a close second to the routines scheduling of Claude.
00:44:16.719 --> 00:44:23.599
I wish Claude would just be able to unify those somehow um and bring some of those better features in routines into their schedule tasks.
00:44:23.679 --> 00:44:26.639
But uh, hey, if you split it up, you can still get the power.
00:44:26.719 --> 00:44:32.639
But yeah, unfortunately, it doesn't kind of have that cohesive memory, which I think is so important to wrap your scheduled tasks up in.
00:44:32.719 --> 00:44:37.360
Otherwise, you may spend more time going back and adjusting them and making them better, anyways.
00:44:37.679 --> 00:44:40.880
All right, so what should you schedule as we wrap here?
00:44:41.039 --> 00:44:45.760
Well, I would ask codecs or chat GPT work first, right?
00:44:45.840 --> 00:45:01.199
Uh unfortunately, Claude Code doesn't have cross-chat or cross-cowork code memory, um, but uh unless it commits something to memory, but I would go through and spend time seeing what you actually do manually inside of these platforms.
00:45:01.360 --> 00:45:03.599
From there, find those repetitive tasks.
00:45:03.679 --> 00:45:05.119
That's what you should be scheduling.
00:45:05.280 --> 00:45:18.480
The other thing, kind of a little hack here, is look at your browser history or give it to an agent and see what your most viewed websites are, and then tie those websites to a deliverable and then schedule a task for that, right?
00:45:18.559 --> 00:45:28.400
The biggest thing is being able to do these things that are time intensive, whether it's research for the Chat GPT work hack that I told you, clicking around on a bunch of websites, right?
00:45:28.559 --> 00:45:39.360
See where you're spending your most time, either manually pushing the buttons in Gemini, in Chat GPT, in Claude, turn those into scheduled tasks, or just where are you spending a lot of your time on the internet?
00:45:39.599 --> 00:45:40.880
The internet's a wasteful place.
00:45:40.960 --> 00:45:42.159
Don't spend time there.
00:45:42.320 --> 00:45:47.280
All right, and then last but not least, the other one is start with this simple triage dashboard, right?
00:45:47.360 --> 00:45:49.360
That's updated every single day.
00:45:49.519 --> 00:46:11.280
It's obviously only going to work in Chat GPT and Claude, uh, but you know, bringing in your you know connectors for mail, calendar, Slack, Teams, your CRM, your major KPIs, your most used connectors, MCPs, etc., being able to automatically have an agent go through, pull all the most important things, and essentially have a start here dashboard, right?
00:46:11.360 --> 00:46:13.760
It's a very, you know, simple use case.
00:46:13.920 --> 00:46:21.039
Maybe that's the 2026 version of like the 2024, uh, you know, improve this email, right?
00:46:21.199 --> 00:46:27.199
Probably in 2026, it's essentially having a custom dashboard and triaging everything and telling you where to start.
00:46:27.440 --> 00:46:33.840
But if you are newer to the game and you haven't already done that, I think that's a great place to start with a scheduled task or action.
00:46:34.079 --> 00:46:37.119
All right, that is a wrap, y'all.
00:46:37.199 --> 00:46:46.159
Uh, so remember, if you haven't already, we're gonna be kicking off going back to school with our start here series uh starting Monday.
00:46:46.239 --> 00:46:47.199
So don't be late.
00:46:47.280 --> 00:46:48.159
I'll see you then.
00:46:48.320 --> 00:46:53.760
And if you haven't already, please go to your everydayai.com, sign up for the free daily newsletter.
00:46:53.840 --> 00:46:54.480
Thanks for tuning in.
00:46:54.559 --> 00:46:57.199
We'll see you back tomorrow in every day for more everyday AI.
00:46:57.280 --> 00:46:58.159
Thanks, y'all.
00:46:58.960 --> 00:47:02.239
And that's a wrap for today's edition of Everyday AI.
00:47:02.400 --> 00:47:03.519
Thanks for joining us.
00:47:03.760 --> 00:47:07.519
If you enjoyed this episode, please subscribe and leave us a rating.
00:47:07.599 --> 00:47:09.199
It helps keep us going.
00:47:09.440 --> 00:47:16.559
For a little more AI magic, visit your everydayai.com and sign up to our daily newsletter so you don't get left behind.
00:47:16.719 --> 00:47:19.360
Go break some barriers, and we'll see you next time.
