WEBVTT
00:00:01.439 --> 00:00:06.559
I am the only full-time employee at Everyday AI.
00:00:06.879 --> 00:00:17.359
Sure, I have a few contractors helping out, but I'm the only full-time human responsible for creating this daily podcast that reaches millions of people a year.
00:00:17.600 --> 00:00:28.000
Others are kind of always shocked to hear that, as most of the other top 25 tech podcasts have large teams to produce the same amount that we produce.
00:00:28.320 --> 00:00:30.960
But do you know what I do all day?
00:00:31.199 --> 00:00:32.960
I just manage agents.
00:00:33.200 --> 00:00:38.079
I mean, literally, right now I have codecs running with at least 50 sub-agents active.
00:00:38.240 --> 00:00:45.759
I have codecs controlling Claude Code, and I have scheduled agents running in Google Gemini, Claude Cowork, Copilot, and others.
00:00:46.000 --> 00:00:56.479
Maybe I'm a little overly productive from a single human to output perspective, but it's about to get worse or maybe better depending on your perspective.
00:00:56.640 --> 00:01:09.760
That's because as soon as today or Friday at the latest, we'll have a new open AI model powered by Cerebrus that runs up to 20 times faster than their current models at the same capabilities.
00:01:10.000 --> 00:01:14.400
And that got me thinking, what the heck is the future of work going to be like?
00:01:14.640 --> 00:01:22.719
I mean, I'm already agent pill and producing more content and hopefully value than much larger teams with many more humans.
00:01:22.959 --> 00:01:28.319
So what happens when all of my agents are 20 times faster?
00:01:28.719 --> 00:01:35.920
I'm not sure yet, but I wanted to kind of opine about that out loud on today's show and explore this weird convergence that we're in now.
00:01:36.079 --> 00:01:50.480
Because not only are agents producing economically valuable work end-to-end at a higher quality bar than expert humans, but now we are entering a new phase where those AI agents might be working 20 times faster.
00:01:50.640 --> 00:02:02.719
And I don't know if anyone truly understands what this convergence means for the future of work, but let's unpack it and see what we might come up with, anyways, on today's episode of Everyday AI.
00:02:02.799 --> 00:02:04.640
So here is the big picture.
00:02:04.799 --> 00:02:18.000
What the heck is going to happen when millions of professionals like me and you, business leaders that are on the cutting edge of AI, what's going to happen when we just become managers of faster agents and they become more powerful?
00:02:18.159 --> 00:02:24.560
Uh, right now, AI is fast and parallel enough that one person can direct dozens of agents a day.
00:02:24.719 --> 00:02:26.479
That's pretty much all I do all day.
00:02:26.719 --> 00:02:33.199
But open AI, as well, we are seeing maybe by the time you're listening to this, it will already be released.
00:02:33.360 --> 00:02:41.280
But at least as of now, uh, you know, the rumors are saying that we're going to be getting a new open AI model powered by Cerebrus.
00:02:41.520 --> 00:02:44.240
That's essentially this really fast chip.
00:02:44.319 --> 00:02:54.560
Uh, and essentially, let's say you put um, you know, a very you know, smart agent out there and it thinks, and you know, normally it takes uh five minutes and you get a great output.
00:02:54.719 --> 00:02:57.520
Well, what happens when that five minutes becomes 30 seconds?
00:02:57.599 --> 00:03:02.719
And normally you go, I don't know, manage other agents or do other tasks during that time that you're waiting.
00:03:02.800 --> 00:03:06.159
What happens when things are just instantly done?
00:03:06.319 --> 00:03:08.879
Um, and the upside here is is real, right?
00:03:08.960 --> 00:03:12.080
With workers reporting more autonomy and more outputs.
00:03:12.240 --> 00:03:19.599
But the risk is maybe a workday spent managing machines while human trust and mentorship quietly just thins out.
00:03:19.840 --> 00:03:22.159
So on today's show, here's what you're gonna learn.
00:03:22.319 --> 00:03:28.639
You're gonna know why employees are increasingly asking agents before managers, specialists, or awkward conversations.
00:03:28.800 --> 00:03:34.479
You're gonna understand how faster inference turns solo work into supervising parallel machine teams.
00:03:34.719 --> 00:03:41.039
You're gonna know how winning companies can empower human judgment while still automating most of the manual knowledge work.
00:03:41.199 --> 00:03:47.919
And you're gonna leave the show with the three human handoffs to protect before your workflow quietly deletes them.
00:03:48.159 --> 00:03:49.919
All right, welcome to everyday AI.
00:03:50.000 --> 00:03:52.000
If you're new here, my name's Jordan Wilson.
00:03:52.080 --> 00:03:53.120
We do this every day.
00:03:53.280 --> 00:03:53.919
It's for you.
00:03:54.080 --> 00:04:05.039
It's your daily, unedited, unscripted live stream podcast, and free daily newsletter helping business leaders like you and me keep up with drinking from the fire hose of new AI updates.
00:04:05.360 --> 00:04:12.159
I turn that fire hose down, I serve you a nice glass of water and say, here, this is what matters today.
00:04:12.319 --> 00:04:16.160
Go take this information and share it with your team and uh become really smart in AI.
00:04:16.240 --> 00:04:17.360
So that's what we do here.
00:04:17.519 --> 00:04:19.120
Hopefully, the podcast is helpful.
00:04:19.279 --> 00:04:22.720
If you didn't know, we recap it all in our daily newsletter.
00:04:22.800 --> 00:04:27.519
So make sure that you go check out our newsletter for that and all of the other AI news.
00:04:27.920 --> 00:04:28.480
All right.
00:04:28.800 --> 00:04:34.160
So when did this start happening?
00:04:34.399 --> 00:04:36.399
Um, I'm not sure.
00:04:37.360 --> 00:04:44.720
As someone that follows this every single day, this kind of convergence now that is maybe gonna start today for people on the edge of the edge.
00:04:44.800 --> 00:04:49.199
Uh, but I think the rest of the enterprise will start to feel with this in maybe six to nine months.
00:04:49.439 --> 00:04:58.480
But we are now gonna be at this convergence very quickly, um, where we've had these extremely capable agents now for the most part running on desktop.
00:04:58.639 --> 00:05:03.680
So whether you're using Codex, ChatGPT work, Claude Desktop, something else, right?
00:05:03.759 --> 00:05:10.480
But I'd say those and maybe Cursor are the big players, at least when it comes to people, number of people using them.
00:05:10.639 --> 00:05:15.839
I don't think Google is there just yet uh with you know anti-gravity or their Gemini desktop.
00:05:15.920 --> 00:05:21.839
Um, Microsoft CEO Satya Nadella did mention again uh yesterday their super app coming out.
00:05:22.000 --> 00:05:28.639
So I do know that most of our uh many of our listeners out there are using Microsoft Windows and Copilot.
00:05:28.879 --> 00:05:34.639
So you'll be living in this world soon too, if you haven't already, maybe with um, you know, co-pilot cowork.
00:05:34.800 --> 00:05:47.439
But we're at this place now where we have these agents that if you know what you're doing, they will literally on demand um output the same level, the same asset, the same deliverable as expert humans, right?
00:05:47.519 --> 00:05:58.319
We see this in uh important benchmarks like GDP Val, that, you know, um a human and a large language model get the same inputs and then they create outputs.
00:05:58.399 --> 00:06:01.199
They're judged blindly by subject matter experts.
00:06:01.360 --> 00:06:06.480
And, you know, we're at the point now where AI models win or tie about 90% of the time.
00:06:06.720 --> 00:06:18.079
So AI models or AI agents can actually do the same work that humans do and create these assets or artifacts or deliverables, uh, which is cool and all, right?
00:06:18.240 --> 00:06:23.120
But you know, I'd say the downside, no, it's not even technically a downside, right?
00:06:23.199 --> 00:06:25.040
I always tell people be a little patient.
00:06:25.120 --> 00:06:27.600
Um, but you have to number one know what you're doing.
00:06:27.759 --> 00:06:38.160
Uh, number two, you have to um understand uh kind of agent guardrails and context engineering, which those are kind of basics uh table uh uh table stakes now.
00:06:38.240 --> 00:06:40.399
But number three is a little bit of time, right?
00:06:40.480 --> 00:06:43.439
And that little bit of time is now what could be changing today.
00:06:43.600 --> 00:06:50.959
Whereas before, you know, I have agents that work for many hours, but I'd say my average agent task maybe lasts about five minutes.
00:06:51.120 --> 00:06:54.319
So for the most part, all day I'm just flipping between agents, right?
00:06:54.480 --> 00:06:58.480
And one of the big things I like doing is doing that on uh ChatGPT voice mode.
00:06:59.040 --> 00:07:08.560
Um but what happens then when that five minutes uh becomes 30 seconds, right?
00:07:09.120 --> 00:07:17.600
I think of so much of what I actually accomplish uh day to day, it's when I'm waiting for agents, I go check in on other agents.
00:07:17.759 --> 00:07:27.680
So then at least, you know, by the time the day is done, right, in theory, I've had time to check in and look at those artifacts and look at those outputs and you know, manually approve them.
00:07:28.079 --> 00:07:29.839
So what happens when things go faster?
00:07:30.079 --> 00:07:38.959
Well, one thing is I I think where we're gonna start is workers are just gonna start asking agents now, right?
00:07:39.199 --> 00:07:51.439
Because not only is this convergence of these two things happening faster and smarter agents, um, but also that takes away, I think ultimately human connection.
00:07:51.600 --> 00:07:55.040
Um, and I think where, you know, people for the last 20 years, right?
00:07:55.199 --> 00:08:03.839
Maybe um I've been guilty of this managing people in the past, and maybe you said this or it's happened to you, but you know, saying, hey, did you Google that?
00:08:04.000 --> 00:08:08.879
Um, right now one of the things is gonna be like, hey, did you ask our agents on that?
00:08:08.959 --> 00:08:13.360
Um, which isn't necessarily a bad thing in theory, right?
00:08:13.680 --> 00:08:19.040
Um but it's gonna take away human connection, right?
00:08:19.279 --> 00:08:20.079
Even for me, right?
00:08:20.160 --> 00:08:24.959
I started the show by saying I'm the only employee at everyday, the only full-time employee at everyday AI.
00:08:25.120 --> 00:08:32.080
I'm not saying that's a good thing or a bad thing, but I'm not talking to a lot of humans all day, right?
00:08:32.159 --> 00:08:34.799
For the most part, I'm talking with agents.
00:08:35.919 --> 00:08:39.360
Um again, pros and cons, good and bad.
00:08:39.440 --> 00:08:49.039
I'm able to produce a lot more, but maybe I'm getting fewer outside perspectives that aren't tied to my own custom instructions and dynamic data that I'm feeding these models.
00:08:49.279 --> 00:09:04.320
Um, but right now, studies show uh so Adobe just came out with uh an interesting study this month that showed that 68% of people um ask agents obvious questions, and only 4% of uh those people would ask their managers.
00:09:04.559 --> 00:09:17.120
So people are like more than 15 times um 16 times more likely to ask an agent something than ask their manager.
00:09:17.360 --> 00:09:20.559
Again, not saying that's a bad thing, but what does that do to human connection?
00:09:20.639 --> 00:09:26.240
And workers chose the agent because it answers instantly and never makes them feel unprepared or dumb.
00:09:26.399 --> 00:09:32.559
And then once that agent answers first, workers kind of stop routing the work to specialists too, because then they build more trust.
00:09:32.639 --> 00:09:36.799
And they're like, okay, well, I was gonna ask my manager on this, but the agent did great.
00:09:36.879 --> 00:09:41.440
And I was gonna take this over to the product team, but I asked our agent and it that did great too, right?
00:09:41.519 --> 00:09:50.960
So, you know, you start to be able to build faster in silos uh versus maybe building more intentionally at a slower pace, but in a collaborative fashion.
00:09:51.279 --> 00:10:01.039
So that's the other thing, is now all of a sudden we're getting a lot more skills that we didn't have before.
00:10:01.200 --> 00:10:07.200
Um, you know, and we're able to attempt work that we couldn't previously because we have a new set of skills.
00:10:07.360 --> 00:10:10.480
You know, kind of the the best way for me to look at this, right?
00:10:10.559 --> 00:10:16.159
Way back in the day, um, you know, I used to have a marketing and advertising agency, uh, right.
00:10:16.240 --> 00:10:20.799
You know, kind of a big part of my background is, you know, I was a writer, I was a journalist before.
00:10:20.960 --> 00:10:29.039
Um, and you know, as a marketer, sometimes you have to wear mini hats, but at a bigger company, a lot of times it's you have people who are specialists, right?
00:10:29.120 --> 00:10:35.360
You have people who just maybe write Google ads, and that's all they've done for, you know, 20 years.
00:10:35.519 --> 00:10:48.559
Um, you know, now it's getting to the point where one marketer, um, maybe they can't do it at an A rating, but they could maybe do 20 jobs at a B rating, whereas before maybe they only had one specialty, right?
00:10:48.639 --> 00:11:08.320
So now um as we get collectively better in the enterprise at sharing our um our skill sets, uh uh recording and um distributing kind of our our nuanced understanding um and kind of internal IP, so to speak, and give that to others via shared agents or shared skills, right?
00:11:08.480 --> 00:11:19.759
All of a sudden, individual humans don't always need well their managers' approval or uh even a specialist to help complete the work.
00:11:20.960 --> 00:11:27.279
And every handoff, right, those handoffs, the human to human handoff that we're starting to decrease.
00:11:27.440 --> 00:11:30.399
Um, it deletes a human conversation.
00:11:30.480 --> 00:11:33.120
Um, it deletes mentorship, uh, right.
00:11:33.200 --> 00:11:40.080
Uh handoff is kind of once something is carried through explanation, correction, introduction, and a moment of shared context.
00:11:40.159 --> 00:11:52.240
So it was purplies, I believe that's how it's pronounced, purplies, uh, May 2026 study that found that 63% of people used AI to avoid a difficult workplace conversation, right?
00:11:52.480 --> 00:11:57.919
Um, and and that's you know, that one skipped conversation multiplies.
00:11:58.080 --> 00:12:07.759
Um, and I think you also start to get internal project drift there, but I think so much of even I'm thinking of where I am today, right?
00:12:07.919 --> 00:12:19.759
Probably the the reason maybe why I've been able to have whatever level of success that I have with this weird little podcast thing, um, is you know, I've had a lot of great mentors um throughout the years.
00:12:19.840 --> 00:12:29.440
I've been able to learn from very smart people, and sometimes that comes from conversations when things go awry, um, you know, or when things go great and having that be a teachable moment.
00:12:29.600 --> 00:12:40.159
So I'm just thinking, you know, yes, uh, there's you know, working with agents, you can get all those things and build them in, but you know, agents by default are sometimes overly sycophantic.
00:12:40.240 --> 00:12:56.080
Um, you know, and then I think you can almost develop a certain uh AI psychosis, you know, in the good way or bad way, uh, versus, well, when you had a team of humans and collaborating more with humans, I don't think that that was as big of a problem, right?
00:12:56.159 --> 00:13:08.000
Working in these silos that, yes, maybe you are, you know, two, three, four, five times more productive in terms of the um business value that you're able to create, but at what cost?
00:13:08.879 --> 00:13:16.639
So the translation here is well, we have more collective skills that we didn't have before, but and we also have fewer productivity roadblocks, right?
00:13:16.720 --> 00:13:18.080
The two things that I just covered.
00:13:18.240 --> 00:13:21.679
And well, that gives us all the reason to be more productive.
00:13:21.919 --> 00:13:24.960
And well, things are getting faster now, too.
00:13:25.120 --> 00:13:34.639
Um, that's because uh, you know, today, maybe tomorrow, but uh OpenAI said in July that they're going to release the new GPT-5.
00:13:41.039 --> 00:13:46.000
So if you don't speak tokens, uh, that's about 20 times faster than the current models, right?
00:13:46.080 --> 00:14:03.200
So if you go into you know codex like like I do, and you know, you turn on you know GPT-5.6 soul and you put it on, you know, extra high or high or ultra or something like that, even if you're throwing a hard task, you know, you might only have to wait five, six, seven minutes.
00:14:03.360 --> 00:14:05.600
Now all of a sudden that's gonna be a matter of seconds.
00:14:05.679 --> 00:14:12.720
So the value there is it's the throughput, it's with the research, the drafting analysis and review all running simultaneously.
00:14:12.879 --> 00:14:15.840
So a slow agent is a tool that you wait on.
00:14:16.000 --> 00:14:23.120
And I don't think that's necessarily a bad thing because what I have always done when I'm waiting on agents, I'm usually reading the chain of thought.
00:14:23.279 --> 00:14:24.159
Um, right.
00:14:24.399 --> 00:14:27.759
If in in that five minutes, I'm literally looking at the chain of thought.
00:14:27.919 --> 00:14:32.879
I'm making sure that it's calling the right tools, I'm making sure it's pulling in all the correct dynamic data.
00:14:32.960 --> 00:14:34.559
Because it's like, what else am I gonna do?
00:14:34.720 --> 00:14:41.759
Well, yeah, probably multitask with some uh some different agents and go in and check on my blue lights that you know require my input.
00:14:41.840 --> 00:14:46.879
But ultimately, I'm spending more time do it being the expert-driven loop, right?
00:14:47.120 --> 00:14:51.440
So I think that this is actually could be a big problem, right?
00:14:51.519 --> 00:14:59.200
When AI is almost too fast, even the people that are maybe trying to be responsible, uh, right, all of a sudden everyone knows.
00:14:59.279 --> 00:15:01.200
I mean, you know, in a year or so, everyone's gonna know.
00:15:01.360 --> 00:15:04.720
Oh, the AI is gonna be way more powerful and it's gonna be even faster.
00:15:04.960 --> 00:15:09.840
Whereas right now, if you're using AI in the right way, in theory, right, you're getting more done.
00:15:09.919 --> 00:15:15.039
But in you, it's almost like you're going slower, but way faster.
00:15:15.679 --> 00:15:21.200
You know, it's like you're waiting, but you're accomplishing more, right?
00:15:21.279 --> 00:15:31.679
So instead of working on how we would traditionally do it, you know, in pre-AI days, you know, me, one human, I would work on one project across these 20 little steps.
00:15:31.759 --> 00:15:38.960
Now, conversely, I'm working on 20 little projects, and my only one step is just checking in on what the agents are doing, right?
00:15:39.039 --> 00:15:54.000
But uh, you know, usually I stagger it in a way where I can um understand, um, kind of you know, prioritize um explainability and and you know, making sure the guardrails that I've set are being followed, making sure my agents aren't crashing, all of these things.
00:15:54.080 --> 00:15:55.519
But what happens when this is done?
00:15:55.759 --> 00:15:57.279
Am I still gonna go and do that?
00:15:57.440 --> 00:15:58.159
I hope so.
00:15:58.320 --> 00:16:07.360
But when things are done faster and when the models are smarter, won't my human nature just be to be like, well, I should go grow the business in a new way.
00:16:07.519 --> 00:16:09.679
I should go get more and more things done.
00:16:09.840 --> 00:16:10.480
Um, right.
00:16:10.639 --> 00:16:15.279
A fast just team just now needs more management.
00:16:15.919 --> 00:16:20.320
So directing agents is just middle management without the manager title, right?
00:16:20.399 --> 00:16:37.039
I do see the future um where most uh you know, most knowledge uh, you know, subject matter experts, most people, if if you're a knowledge worker, you sit in front of a computer um and you create value for your business, which is what many of us do.
00:16:37.200 --> 00:16:40.960
Um, I think people for the most part, you're gonna be like a middle manager, right?
00:16:41.120 --> 00:16:42.480
Of just agents.
00:16:43.360 --> 00:16:49.919
I like I I actually think that uh in theory, the middle management layer is going to get bigger and thicker.
00:16:50.000 --> 00:16:52.720
Um, I think there's gonna be fewer people at the entry level.
00:16:52.799 --> 00:16:57.600
I think there's gonna be fewer people in the, you know, the VP, C suite director level.
00:16:57.759 --> 00:17:04.160
I think for the most part, you know, your average knowledge worker, that middle uh middle management tier, right?
00:17:04.240 --> 00:17:12.160
When we think it's not really needed, well, I think true, you know, if we look at middle management today in an AI native world, it's not needed.
00:17:12.240 --> 00:17:15.759
But I think just everyone else is gonna become kind of that middle layer, right?
00:17:15.920 --> 00:17:18.559
People are gonna be managing more agents, not people.
00:17:18.720 --> 00:17:22.559
And yes, there's still gonna be, you know, humans reporting up the food chain.
00:17:22.720 --> 00:17:30.880
Um, but I think for the most part, the overwhelming majority of work is actually gonna get done in that middle management layer, where traditionally that's not where the work gets done, right?
00:17:31.039 --> 00:17:32.319
They're just managing people.
00:17:32.480 --> 00:17:35.599
Um, so it's I I think that's ultimately what's gonna happen.
00:17:35.839 --> 00:17:43.359
But I think it's normal in AI native companies for decision makers to just publish an agent to help to speed up productivity, right?
00:17:43.519 --> 00:17:59.759
You've seen these uh, you know, stories where literally, you know, a CEO or a VP or you know, someone that manages a big department at a huge company is just literally unloading their entire brain and decision-making process and all their documents and all of these things into an agent, right?
00:17:59.839 --> 00:18:09.119
Let's say it's it's you know, marketing Jane at a Fortune 500 company, you know, Jane just you know dumps her brain, her decision-making process, everything into an agent.
00:18:09.200 --> 00:18:14.240
And the you know, 500 people in marketing, well, they can just get answers straight from the Jane agent.
00:18:14.400 --> 00:18:15.680
They don't necessarily have to wait.
00:18:15.839 --> 00:18:19.519
So the work, um, it just becomes faster, but it becomes four verbs.
00:18:19.599 --> 00:18:26.559
I think you decide what matters, you delegate, you inspect the exceptions, and then you integrate the results, right?
00:18:26.720 --> 00:18:31.279
So, but those things can happen now like agents can in parallel.
00:18:31.440 --> 00:18:37.759
And you have to push that pattern to its limit, and that one person just directs more hours than exist, right?
00:18:37.839 --> 00:18:48.640
That's that's where I find myself and I found myself, you know, since uh kind of uh Claude Cowork um and Claude Code kind of kicked off this phase in late 2025, early 2026.
00:18:48.720 --> 00:18:53.039
And I think Codex um has really took in uh taken over the narrative ever since.
00:18:53.119 --> 00:18:59.039
But you know, essentially I'm directing more um hours than exist in a day, right?
00:18:59.119 --> 00:19:03.599
I'm literally getting done what used to take me even two years ago, right?
00:19:03.680 --> 00:19:11.839
Even in the early Chat GPT days, I'm getting uh about 30 hours of 2024 me work done every single day.
00:19:12.000 --> 00:19:17.599
So it's just you're directing more hours, more traffic than can fit on a traditional runway.
00:19:17.839 --> 00:19:23.359
Um, and you know, in uh OpenAI report came out with this, they showed how their team was using codex.
00:19:23.599 --> 00:19:32.720
And uh what I found interesting was uh OpenAI staff with a 99th percentile generated over 60 hours of agent turns every single day.
00:19:32.880 --> 00:19:35.039
So, yeah, that's kind of the anomaly, right?
00:19:35.119 --> 00:19:46.000
That's not the average because they're building this technology, but I think eventually that is going to become the default, and that is gonna force people, I think, to question how do you work, right?
00:19:46.160 --> 00:19:48.720
What is the value of my actual time?
00:19:49.039 --> 00:19:55.119
And how can we responsibly, you know, use these agents as they get even smarter and even faster.
00:19:55.279 --> 00:19:58.480
I actually think we're in a nice little groove where we are today, right?
00:19:58.559 --> 00:20:05.519
There's been all this, you know, talk uh in the last, you know, 36 hours or so of this concept of AI pacing, you know.
00:20:05.759 --> 00:20:19.359
Um uh essentially a lot of researchers, um particularly at OpenAI Anthropic, but also Google, Meta and other labs, you know, kind of sign this letter saying, hey, we support, you know, essentially calling it like pacing, right?
00:20:19.440 --> 00:20:23.519
But it's kind of just means like, hey, it's okay if we slow down.
00:20:23.680 --> 00:20:24.480
Will that happen?
00:20:24.720 --> 00:20:25.599
Probably not.
00:20:25.759 --> 00:20:26.240
Um, right.
00:20:26.400 --> 00:20:30.720
Is it the responsible thing for you know some sort of AI pacing to happen?
00:20:30.960 --> 00:20:31.680
Probably.
00:20:31.839 --> 00:20:36.240
Um, because the issue is the capability gap is growing.
00:20:36.400 --> 00:20:40.079
The model overhang is growing and it's not going to get any slower, right?
00:20:40.240 --> 00:20:50.720
Especially as uh, you know, over the past couple of months, we've gotten, you know, real results uh from the big labs talking about RSI recursive self-improvement, where the models are improving themselves, right?
00:20:50.880 --> 00:20:55.359
Yesterday, OpenAI literally just announced like an 18% improvement via RSI.
00:20:55.519 --> 00:21:01.920
They essentially said, Hey, we got, you know, we use GPT 5.6 to make itself better, and it's like 18% better.
00:21:02.000 --> 00:21:04.559
It's gonna see, you know, 18% cost savings now.
00:21:04.720 --> 00:21:07.759
Um, so this is where it's at.
00:21:07.839 --> 00:21:15.599
These models are gonna get better and better, faster and faster, but way at a much faster accelerated rate than humans can keep up with.
00:21:15.759 --> 00:21:19.839
Um, so yeah, there's this whole pacing thing that maybe we'll tackle at a different time.
00:21:20.079 --> 00:21:29.519
But the the the brutal reality is if you want your team to be able to keep up and get ahead, that's something that you have to tackle right now, head on.
00:21:29.599 --> 00:21:36.319
Uh, some other good stats that recently came out, a workday foundation study in May found that 86% of users felt more productive.
00:21:36.480 --> 00:21:42.799
But those same workers describe the conversations um that rarely moved past transactional work needs, right?
00:21:42.960 --> 00:21:44.240
And that's kind of where I feel right now.
00:21:44.319 --> 00:21:45.920
It's like, yeah, I'm getting way more done.
00:21:46.079 --> 00:21:48.400
The quant the content I hope is good.
00:21:48.559 --> 00:21:49.440
The quality is good, right?
00:21:49.519 --> 00:21:52.160
What I produce is much different than what most people produce.
00:21:52.319 --> 00:21:55.680
Um, but it's just kind of transactional, right?
00:21:55.759 --> 00:21:59.759
I've talked many times on the show about I'm not even.
00:22:00.319 --> 00:22:12.880
Because my you know, the throughput, the output is so insanely high for what I'm doing on a daily basis as an individual full-time human, I don't retain information like I used to because I think so much information now is transactional.
00:22:13.039 --> 00:22:18.720
And maybe that's something that we, you know, the future workforce needs to understand.
00:22:18.799 --> 00:22:26.480
Um, because I think, you know, have you ever thought about how you can learn a concept better if you write it down by hand or if you teach it to someone, right?
00:22:26.640 --> 00:22:30.160
I'm a big believer in you haven't really learned something until you teach, right?
00:22:30.240 --> 00:22:33.200
That's uh one of the reasons why I started a daily podcast.
00:22:33.279 --> 00:22:35.200
I'm like, I really want to learn this AI thing.
00:22:35.359 --> 00:22:39.440
I need to learn it well enough that I can teach at least one person, uh, right.
00:22:39.599 --> 00:22:41.680
Luckily, hopefully it's helped more than one person.
00:22:41.839 --> 00:22:43.839
Um, but I think the same is true.
00:22:44.000 --> 00:22:48.640
Um, because AI, it it does all that hard work for you, right?
00:22:48.720 --> 00:22:53.599
You no longer have to go onto 20 web pages and read it all and be like, ah, this doesn't make sense.
00:22:53.759 --> 00:22:54.799
This does make sense.
00:22:55.039 --> 00:22:56.079
I'm not sure about this.
00:22:56.160 --> 00:22:57.359
Let me look at a little more.
00:22:57.519 --> 00:23:03.440
Now you just put something into whatever AI system you're using, you get a personalized, customized output.
00:23:03.599 --> 00:23:07.759
And well, most people don't even take the time to truly read it and understand it.
00:23:07.920 --> 00:23:15.839
So, you know, there is this uh disconnection uh between you know the productivity and disconnection that's shared in one workday.
00:23:16.000 --> 00:23:19.279
And this study shows that that is the mechanism.
00:23:20.160 --> 00:23:23.680
Um, and I've actually, you know, touched on some of these other concepts.
00:23:23.839 --> 00:23:32.400
So, you know, if if you do want to, you know, dive in a little bit deeper on, you know, some of these uh downsides of using AI the right way.
00:23:32.480 --> 00:23:35.200
I actually had an episode, episode 757.
00:23:35.359 --> 00:23:37.279
Uh, that was part of the start here series.
00:23:37.440 --> 00:23:44.400
It was called The Seven Silent Sins of Doing AI Right, How to Spot and Overcome the Invisible AI work traps.
00:23:44.480 --> 00:23:52.319
And three that I wanted to share that are kind of relevant to what we're talking about today is number one, accidental de-skilling, which is kind of one of the things I just referenced there.
00:23:52.480 --> 00:23:53.680
That's where AI does the work.
00:23:53.839 --> 00:23:55.119
So you lose the skill.
00:23:55.200 --> 00:23:58.000
Uh, I talked about the agent bun sandwich.
00:23:58.079 --> 00:24:01.599
Um, and that's when agents start to hollow out your core expertise.
00:24:01.759 --> 00:24:05.680
And then the compression tax, where the AI speed overloads your brain.
00:24:05.839 --> 00:24:08.559
And that's where I think we might be getting into, right?
00:24:08.640 --> 00:24:11.759
With this um, you know, this cerebrus, right?
00:24:11.839 --> 00:24:15.680
Which might sound like a dorky niche thing, and you're like, okay, why do you keep talking about it?
00:24:15.839 --> 00:24:27.119
Because again, think if you are an AI power user now, and I said, hey, tomorrow, you or your company, your team can accomplish 20x what you accomplished today.
00:24:27.279 --> 00:24:34.240
If you are an agent-pilled team, and I'm like, you can accomplish 20 times more today, that's some compression tax happening right there.
00:24:34.400 --> 00:24:34.640
All right.
00:24:34.720 --> 00:24:38.160
So yeah, make sure uh go check out episode 757.
00:24:38.720 --> 00:24:40.880
So let's be fair to AI, though.
00:24:41.039 --> 00:24:43.359
Some of this stuff, we don't need it, right?
00:24:43.440 --> 00:24:47.680
How many times have you gone through a three-hour meeting and you're like, that could have been a single bullet point?
00:24:47.839 --> 00:24:49.839
That could have been a single Slack message, right?
00:24:50.000 --> 00:24:55.680
Or waiting two days for a basic answer is not, you know, culture or mentorship.
00:24:55.759 --> 00:25:01.680
It's just maybe the signs of an antiquated business hierarchy that shouldn't have existed in the first place, right?
00:25:01.759 --> 00:25:09.039
I think most companies um didn't quickly redesign how they worked um for an AI native workplace.
00:25:09.279 --> 00:25:17.920
So, you know, uh you get to then decide um which collaboration is intentional.
00:25:18.319 --> 00:25:25.119
Um, you know, human to human, you get to choose, um, kind of pick your spots where that happens if you are AI native.
00:25:25.599 --> 00:25:29.039
Okay, so as we wrap up here, I want to try to answer this question.
00:25:29.200 --> 00:25:41.839
How do you strike the balance then of you know the reality of we're gonna be managing more agents, which might lead to less human connection and probably an increased expectation of productivity?
00:25:41.920 --> 00:25:44.799
How do all of those three things uh build together?
00:25:44.960 --> 00:25:47.519
And I know this isn't for 100% of organizations, right?
00:25:47.599 --> 00:25:54.000
I I understand that, but I do think that more organizations, more teams will be shifting toward this than the opposite.
00:25:54.240 --> 00:25:55.759
So, how do you meet that head on?
00:25:55.839 --> 00:26:03.680
Well, I think that winning companies are gonna intentionally automate the manual work, but keep and prioritize human judgment.
00:26:03.920 --> 00:26:07.119
So I think generic human in the loop will always fail.
00:26:07.279 --> 00:26:09.359
So you have to name one expert owner.
00:26:09.599 --> 00:26:21.519
I always talk about expert-driven loops, and expert-driven loops actually require instead of one passive human generalist, it requires active expert human collaboration and communication.
00:26:21.680 --> 00:26:23.039
So I think that's one thing right there.
00:26:23.200 --> 00:26:30.400
But also, I think winning teams um know which friction to remove and which human handoffs to protect and maybe elevate.
00:26:30.640 --> 00:26:32.880
So then how do we strike the balance then?
00:26:33.680 --> 00:26:41.839
Well, I think, yeah, as as my slides are losing, losing my place here.
00:26:42.079 --> 00:26:45.200
Uh sorry, yeah, all all screwed up.
00:26:46.000 --> 00:26:46.640
There we go.
00:26:46.799 --> 00:26:48.319
My slides got all out of order.
00:26:48.480 --> 00:26:53.119
So um, here's the three human handoffs that I think you need to protect.
00:26:53.279 --> 00:27:00.240
So, number one, you need to protect the judgment by naming the human who owns consequences when the answer is wrong.
00:27:00.480 --> 00:27:00.799
Okay.
00:27:01.119 --> 00:27:09.039
Um, number two, you need to protect learning by making people uh explain decisions instead of blindly approving agent output.
00:27:09.119 --> 00:27:24.720
So that's kind of the difference between, like I said, passive, lazy, human in the loop, which will always fail, and the correct way to do it, which is expert-driven loops, which is experts um communicating uh with each other, but also expecting, um inspecting agent outputs.
00:27:24.799 --> 00:27:27.680
And then last, you need to protect your belonging, right?
00:27:27.759 --> 00:27:34.160
I think so much of this, um, so much of what we do in our work, it's about having this sense of belonging.
00:27:34.319 --> 00:27:44.480
It's about, you know, some people, for better or worse, wrap their um, you know, self in um their own meaning is who they are in their job, right?
00:27:44.559 --> 00:27:52.160
But I think you can still protect your belonging by giving coaching, recognition, and unstructured conversation a scheduled place to happen.
00:27:52.400 --> 00:27:52.880
All right.
00:27:53.039 --> 00:27:57.119
So, what's my big takeaway here as we get faster agents that are smarter?
00:27:57.359 --> 00:28:04.640
And maybe we're just managing dozens or hundreds of agents and talking to our human coworkers less, but producing more.
00:28:04.799 --> 00:28:07.440
Well, I think three big takeaways for me.
00:28:07.599 --> 00:28:10.960
Number one, everyone needs to increase their ambition, right?
00:28:11.039 --> 00:28:17.599
Not just saying more prompts, more outputs, but being more intentional with what you can accomplish, right?
00:28:17.759 --> 00:28:22.960
Maybe not just doing 20x of what you were doing yesterday because you can do 20x today.
00:28:23.119 --> 00:28:28.240
Maybe instead it's thinking higher, it's thinking harder, it's just being more ambitious.
00:28:28.400 --> 00:28:32.000
Number two, you always have to unmarry your domain expertise.
00:28:32.240 --> 00:28:38.319
That is the uh, you know, the building blocks of what I've been talking about now for three and a half years, unlearning.
00:28:38.400 --> 00:28:43.359
Um, you have to unmarry your domain expertise because it doesn't really matter necessarily anymore.
00:28:43.519 --> 00:28:49.039
Your domain expertise is really just about that, you know, the human agent bun sandwich, right?
00:28:49.119 --> 00:28:55.519
It's it's making sure that you give it the right context and check it in the right way, but you're not going to be doing the 95% in the middle anymore.
00:28:55.599 --> 00:28:59.519
So you have to unmarry your domain expertise as agents get faster and smarter.
00:28:59.680 --> 00:29:05.200
And then last but not least, you have to be prepared to rebuild monthly instead of yearly.
00:29:05.359 --> 00:29:22.000
Um, I think in the pre-agentic phase of 2024 or early 2025, you know, I was still recommending that people revisit their AI strategy roadmaps quarterly, but I would say even if you did it yearly, you know, in 2024, you weren't gonna be left behind.
00:29:22.079 --> 00:29:22.880
Now you will.
00:29:23.039 --> 00:29:34.720
You have to be prepared to literally uh unlearn and rebuild how your company operates every single month, which sounds like an absolutely asinine concept.
00:29:34.880 --> 00:29:38.000
Because before you only had to do it every five years, right?
00:29:38.160 --> 00:29:40.079
Uh you know, or maybe every 10 years.
00:29:40.160 --> 00:29:52.880
And then, you know, as the internet and you know, web two and social media and all these other things uh started to gain in popularity and how people make decisions and all that, maybe, you know, advertising, marketing, communications, you know, those things changes.
00:29:52.960 --> 00:30:08.000
And so, you know, with digital transformation, I think that shortened um that lifespan of you know how your company works, kind of your uh your company's IP, not just your SOPs, but literally how your company works and runs and thinks and makes decisions.
00:30:08.160 --> 00:30:09.519
You can't do it anymore.
00:30:09.759 --> 00:30:25.279
As crazy as it sounds, I think probably a majority of the time for uh successful companies moving forward are going to spend as much time uh unlearning and rebuilding processes as they are actually doing work the old-fashioned way.
00:30:25.440 --> 00:30:25.759
All right.
00:30:25.920 --> 00:30:28.079
So I hope this episode was helpful.
00:30:28.240 --> 00:30:36.079
Just more of a kind of a zoomed out look of saying, I don't necessarily know what the heck is going on or what the heck is happening next.
00:30:36.240 --> 00:30:54.319
But if you look at the writing on the wall, you have to see and you have to understand agents are getting smarter, agents are getting faster, and we, as people who are AI leaders in our respective organizations, well, we're gonna be working with way more agents and probably fewer humans in many cases.
00:30:54.480 --> 00:30:57.279
Um, so I hope this episode was helpful.
00:30:57.359 --> 00:30:58.240
If so, let me know.
00:30:58.319 --> 00:30:59.680
Uh, please subscribe to the podcast.
00:30:59.759 --> 00:31:06.319
If you want to hear more or less of shows like this, uh, do let me know by signing up for the newsletter at your everydayai.com.
00:31:06.400 --> 00:31:07.359
So thank you for tuning in.
00:31:07.519 --> 00:31:10.240
Hope to see you back tomorrow and every day for more everyday AI.
00:31:10.319 --> 00:31:11.359
Thanks, y'all.
00:00:01.439 --> 00:00:06.559
I am the only full-time employee at Everyday AI.
00:00:06.879 --> 00:00:17.359
Sure, I have a few contractors helping out, but I'm the only full-time human responsible for creating this daily podcast that reaches millions of people a year.
00:00:17.600 --> 00:00:28.000
Others are kind of always shocked to hear that, as most of the other top 25 tech podcasts have large teams to produce the same amount that we produce.
00:00:28.320 --> 00:00:30.960
But do you know what I do all day?
00:00:31.199 --> 00:00:32.960
I just manage agents.
00:00:33.200 --> 00:00:38.079
I mean, literally, right now I have codecs running with at least 50 sub-agents active.
00:00:38.240 --> 00:00:45.759
I have codecs controlling Claude Code, and I have scheduled agents running in Google Gemini, Claude Cowork, Copilot, and others.
00:00:46.000 --> 00:00:56.479
Maybe I'm a little overly productive from a single human to output perspective, but it's about to get worse or maybe better depending on your perspective.
00:00:56.640 --> 00:01:09.760
That's because as soon as today or Friday at the latest, we'll have a new open AI model powered by Cerebrus that runs up to 20 times faster than their current models at the same capabilities.
00:01:10.000 --> 00:01:14.400
And that got me thinking, what the heck is the future of work going to be like?
00:01:14.640 --> 00:01:22.719
I mean, I'm already agent pill and producing more content and hopefully value than much larger teams with many more humans.
00:01:22.959 --> 00:01:28.319
So what happens when all of my agents are 20 times faster?
00:01:28.719 --> 00:01:35.920
I'm not sure yet, but I wanted to kind of opine about that out loud on today's show and explore this weird convergence that we're in now.
00:01:36.079 --> 00:01:50.480
Because not only are agents producing economically valuable work end-to-end at a higher quality bar than expert humans, but now we are entering a new phase where those AI agents might be working 20 times faster.
00:01:50.640 --> 00:02:02.719
And I don't know if anyone truly understands what this convergence means for the future of work, but let's unpack it and see what we might come up with, anyways, on today's episode of Everyday AI.
00:02:02.799 --> 00:02:04.640
So here is the big picture.
00:02:04.799 --> 00:02:18.000
What the heck is going to happen when millions of professionals like me and you, business leaders that are on the cutting edge of AI, what's going to happen when we just become managers of faster agents and they become more powerful?
00:02:18.159 --> 00:02:24.560
Uh, right now, AI is fast and parallel enough that one person can direct dozens of agents a day.
00:02:24.719 --> 00:02:26.479
That's pretty much all I do all day.
00:02:26.719 --> 00:02:33.199
But open AI, as well, we are seeing maybe by the time you're listening to this, it will already be released.
00:02:33.360 --> 00:02:41.280
But at least as of now, uh, you know, the rumors are saying that we're going to be getting a new open AI model powered by Cerebrus.
00:02:41.520 --> 00:02:44.240
That's essentially this really fast chip.
00:02:44.319 --> 00:02:54.560
Uh, and essentially, let's say you put um, you know, a very you know, smart agent out there and it thinks, and you know, normally it takes uh five minutes and you get a great output.
00:02:54.719 --> 00:02:57.520
Well, what happens when that five minutes becomes 30 seconds?
00:02:57.599 --> 00:03:02.719
And normally you go, I don't know, manage other agents or do other tasks during that time that you're waiting.
00:03:02.800 --> 00:03:06.159
What happens when things are just instantly done?
00:03:06.319 --> 00:03:08.879
Um, and the upside here is is real, right?
00:03:08.960 --> 00:03:12.080
With workers reporting more autonomy and more outputs.
00:03:12.240 --> 00:03:19.599
But the risk is maybe a workday spent managing machines while human trust and mentorship quietly just thins out.
00:03:19.840 --> 00:03:22.159
So on today's show, here's what you're gonna learn.
00:03:22.319 --> 00:03:28.639
You're gonna know why employees are increasingly asking agents before managers, specialists, or awkward conversations.
00:03:28.800 --> 00:03:34.479
You're gonna understand how faster inference turns solo work into supervising parallel machine teams.
00:03:34.719 --> 00:03:41.039
You're gonna know how winning companies can empower human judgment while still automating most of the manual knowledge work.
00:03:41.199 --> 00:03:47.919
And you're gonna leave the show with the three human handoffs to protect before your workflow quietly deletes them.
00:03:48.159 --> 00:03:49.919
All right, welcome to everyday AI.
00:03:50.000 --> 00:03:52.000
If you're new here, my name's Jordan Wilson.
00:03:52.080 --> 00:03:53.120
We do this every day.
00:03:53.280 --> 00:03:53.919
It's for you.
00:03:54.080 --> 00:04:05.039
It's your daily, unedited, unscripted live stream podcast, and free daily newsletter helping business leaders like you and me keep up with drinking from the fire hose of new AI updates.
00:04:05.360 --> 00:04:12.159
I turn that fire hose down, I serve you a nice glass of water and say, here, this is what matters today.
00:04:12.319 --> 00:04:16.160
Go take this information and share it with your team and uh become really smart in AI.
00:04:16.240 --> 00:04:17.360
So that's what we do here.
00:04:17.519 --> 00:04:19.120
Hopefully, the podcast is helpful.
00:04:19.279 --> 00:04:22.720
If you didn't know, we recap it all in our daily newsletter.
00:04:22.800 --> 00:04:27.519
So make sure that you go check out our newsletter for that and all of the other AI news.
00:04:27.920 --> 00:04:28.480
All right.
00:04:28.800 --> 00:04:34.160
So when did this start happening?
00:04:34.399 --> 00:04:36.399
Um, I'm not sure.
00:04:37.360 --> 00:04:44.720
As someone that follows this every single day, this kind of convergence now that is maybe gonna start today for people on the edge of the edge.
00:04:44.800 --> 00:04:49.199
Uh, but I think the rest of the enterprise will start to feel with this in maybe six to nine months.
00:04:49.439 --> 00:04:58.480
But we are now gonna be at this convergence very quickly, um, where we've had these extremely capable agents now for the most part running on desktop.
00:04:58.639 --> 00:05:03.680
So whether you're using Codex, ChatGPT work, Claude Desktop, something else, right?
00:05:03.759 --> 00:05:10.480
But I'd say those and maybe Cursor are the big players, at least when it comes to people, number of people using them.
00:05:10.639 --> 00:05:15.839
I don't think Google is there just yet uh with you know anti-gravity or their Gemini desktop.
00:05:15.920 --> 00:05:21.839
Um, Microsoft CEO Satya Nadella did mention again uh yesterday their super app coming out.
00:05:22.000 --> 00:05:28.639
So I do know that most of our uh many of our listeners out there are using Microsoft Windows and Copilot.
00:05:28.879 --> 00:05:34.639
So you'll be living in this world soon too, if you haven't already, maybe with um, you know, co-pilot cowork.
00:05:34.800 --> 00:05:47.439
But we're at this place now where we have these agents that if you know what you're doing, they will literally on demand um output the same level, the same asset, the same deliverable as expert humans, right?
00:05:47.519 --> 00:05:58.319
We see this in uh important benchmarks like GDP Val, that, you know, um a human and a large language model get the same inputs and then they create outputs.
00:05:58.399 --> 00:06:01.199
They're judged blindly by subject matter experts.
00:06:01.360 --> 00:06:06.480
And, you know, we're at the point now where AI models win or tie about 90% of the time.
00:06:06.720 --> 00:06:18.079
So AI models or AI agents can actually do the same work that humans do and create these assets or artifacts or deliverables, uh, which is cool and all, right?
00:06:18.240 --> 00:06:23.120
But you know, I'd say the downside, no, it's not even technically a downside, right?
00:06:23.199 --> 00:06:25.040
I always tell people be a little patient.
00:06:25.120 --> 00:06:27.600
Um, but you have to number one know what you're doing.
00:06:27.759 --> 00:06:38.160
Uh, number two, you have to um understand uh kind of agent guardrails and context engineering, which those are kind of basics uh table uh uh table stakes now.
00:06:38.240 --> 00:06:40.399
But number three is a little bit of time, right?
00:06:40.480 --> 00:06:43.439
And that little bit of time is now what could be changing today.
00:06:43.600 --> 00:06:50.959
Whereas before, you know, I have agents that work for many hours, but I'd say my average agent task maybe lasts about five minutes.
00:06:51.120 --> 00:06:54.319
So for the most part, all day I'm just flipping between agents, right?
00:06:54.480 --> 00:06:58.480
And one of the big things I like doing is doing that on uh ChatGPT voice mode.
00:06:59.040 --> 00:07:08.560
Um but what happens then when that five minutes uh becomes 30 seconds, right?
00:07:09.120 --> 00:07:17.600
I think of so much of what I actually accomplish uh day to day, it's when I'm waiting for agents, I go check in on other agents.
00:07:17.759 --> 00:07:27.680
So then at least, you know, by the time the day is done, right, in theory, I've had time to check in and look at those artifacts and look at those outputs and you know, manually approve them.
00:07:28.079 --> 00:07:29.839
So what happens when things go faster?
00:07:30.079 --> 00:07:38.959
Well, one thing is I I think where we're gonna start is workers are just gonna start asking agents now, right?
00:07:39.199 --> 00:07:51.439
Because not only is this convergence of these two things happening faster and smarter agents, um, but also that takes away, I think ultimately human connection.
00:07:51.600 --> 00:07:55.040
Um, and I think where, you know, people for the last 20 years, right?
00:07:55.199 --> 00:08:03.839
Maybe um I've been guilty of this managing people in the past, and maybe you said this or it's happened to you, but you know, saying, hey, did you Google that?
00:08:04.000 --> 00:08:08.879
Um, right now one of the things is gonna be like, hey, did you ask our agents on that?
00:08:08.959 --> 00:08:13.360
Um, which isn't necessarily a bad thing in theory, right?
00:08:13.680 --> 00:08:19.040
Um but it's gonna take away human connection, right?
00:08:19.279 --> 00:08:20.079
Even for me, right?
00:08:20.160 --> 00:08:24.959
I started the show by saying I'm the only employee at everyday, the only full-time employee at everyday AI.
00:08:25.120 --> 00:08:32.080
I'm not saying that's a good thing or a bad thing, but I'm not talking to a lot of humans all day, right?
00:08:32.159 --> 00:08:34.799
For the most part, I'm talking with agents.
00:08:35.919 --> 00:08:39.360
Um again, pros and cons, good and bad.
00:08:39.440 --> 00:08:49.039
I'm able to produce a lot more, but maybe I'm getting fewer outside perspectives that aren't tied to my own custom instructions and dynamic data that I'm feeding these models.
00:08:49.279 --> 00:09:04.320
Um, but right now, studies show uh so Adobe just came out with uh an interesting study this month that showed that 68% of people um ask agents obvious questions, and only 4% of uh those people would ask their managers.
00:09:04.559 --> 00:09:17.120
So people are like more than 15 times um 16 times more likely to ask an agent something than ask their manager.
00:09:17.360 --> 00:09:20.559
Again, not saying that's a bad thing, but what does that do to human connection?
00:09:20.639 --> 00:09:26.240
And workers chose the agent because it answers instantly and never makes them feel unprepared or dumb.
00:09:26.399 --> 00:09:32.559
And then once that agent answers first, workers kind of stop routing the work to specialists too, because then they build more trust.
00:09:32.639 --> 00:09:36.799
And they're like, okay, well, I was gonna ask my manager on this, but the agent did great.
00:09:36.879 --> 00:09:41.440
And I was gonna take this over to the product team, but I asked our agent and it that did great too, right?
00:09:41.519 --> 00:09:50.960
So, you know, you start to be able to build faster in silos uh versus maybe building more intentionally at a slower pace, but in a collaborative fashion.
00:09:51.279 --> 00:10:01.039
So that's the other thing, is now all of a sudden we're getting a lot more skills that we didn't have before.
00:10:01.200 --> 00:10:07.200
Um, you know, and we're able to attempt work that we couldn't previously because we have a new set of skills.
00:10:07.360 --> 00:10:10.480
You know, kind of the the best way for me to look at this, right?
00:10:10.559 --> 00:10:16.159
Way back in the day, um, you know, I used to have a marketing and advertising agency, uh, right.
00:10:16.240 --> 00:10:20.799
You know, kind of a big part of my background is, you know, I was a writer, I was a journalist before.
00:10:20.960 --> 00:10:29.039
Um, and you know, as a marketer, sometimes you have to wear mini hats, but at a bigger company, a lot of times it's you have people who are specialists, right?
00:10:29.120 --> 00:10:35.360
You have people who just maybe write Google ads, and that's all they've done for, you know, 20 years.
00:10:35.519 --> 00:10:48.559
Um, you know, now it's getting to the point where one marketer, um, maybe they can't do it at an A rating, but they could maybe do 20 jobs at a B rating, whereas before maybe they only had one specialty, right?
00:10:48.639 --> 00:11:08.320
So now um as we get collectively better in the enterprise at sharing our um our skill sets, uh uh recording and um distributing kind of our our nuanced understanding um and kind of internal IP, so to speak, and give that to others via shared agents or shared skills, right?
00:11:08.480 --> 00:11:19.759
All of a sudden, individual humans don't always need well their managers' approval or uh even a specialist to help complete the work.
00:11:20.960 --> 00:11:27.279
And every handoff, right, those handoffs, the human to human handoff that we're starting to decrease.
00:11:27.440 --> 00:11:30.399
Um, it deletes a human conversation.
00:11:30.480 --> 00:11:33.120
Um, it deletes mentorship, uh, right.
00:11:33.200 --> 00:11:40.080
Uh handoff is kind of once something is carried through explanation, correction, introduction, and a moment of shared context.
00:11:40.159 --> 00:11:52.240
So it was purplies, I believe that's how it's pronounced, purplies, uh, May 2026 study that found that 63% of people used AI to avoid a difficult workplace conversation, right?
00:11:52.480 --> 00:11:57.919
Um, and and that's you know, that one skipped conversation multiplies.
00:11:58.080 --> 00:12:07.759
Um, and I think you also start to get internal project drift there, but I think so much of even I'm thinking of where I am today, right?
00:12:07.919 --> 00:12:19.759
Probably the the reason maybe why I've been able to have whatever level of success that I have with this weird little podcast thing, um, is you know, I've had a lot of great mentors um throughout the years.
00:12:19.840 --> 00:12:29.440
I've been able to learn from very smart people, and sometimes that comes from conversations when things go awry, um, you know, or when things go great and having that be a teachable moment.
00:12:29.600 --> 00:12:40.159
So I'm just thinking, you know, yes, uh, there's you know, working with agents, you can get all those things and build them in, but you know, agents by default are sometimes overly sycophantic.
00:12:40.240 --> 00:12:56.080
Um, you know, and then I think you can almost develop a certain uh AI psychosis, you know, in the good way or bad way, uh, versus, well, when you had a team of humans and collaborating more with humans, I don't think that that was as big of a problem, right?
00:12:56.159 --> 00:13:08.000
Working in these silos that, yes, maybe you are, you know, two, three, four, five times more productive in terms of the um business value that you're able to create, but at what cost?
00:13:08.879 --> 00:13:16.639
So the translation here is well, we have more collective skills that we didn't have before, but and we also have fewer productivity roadblocks, right?
00:13:16.720 --> 00:13:18.080
The two things that I just covered.
00:13:18.240 --> 00:13:21.679
And well, that gives us all the reason to be more productive.
00:13:21.919 --> 00:13:24.960
And well, things are getting faster now, too.
00:13:25.120 --> 00:13:34.639
Um, that's because uh, you know, today, maybe tomorrow, but uh OpenAI said in July that they're going to release the new GPT-5.
00:13:41.039 --> 00:13:46.000
So if you don't speak tokens, uh, that's about 20 times faster than the current models, right?
00:13:46.080 --> 00:14:03.200
So if you go into you know codex like like I do, and you know, you turn on you know GPT-5.6 soul and you put it on, you know, extra high or high or ultra or something like that, even if you're throwing a hard task, you know, you might only have to wait five, six, seven minutes.
00:14:03.360 --> 00:14:05.600
Now all of a sudden that's gonna be a matter of seconds.
00:14:05.679 --> 00:14:12.720
So the value there is it's the throughput, it's with the research, the drafting analysis and review all running simultaneously.
00:14:12.879 --> 00:14:15.840
So a slow agent is a tool that you wait on.
00:14:16.000 --> 00:14:23.120
And I don't think that's necessarily a bad thing because what I have always done when I'm waiting on agents, I'm usually reading the chain of thought.
00:14:23.279 --> 00:14:24.159
Um, right.
00:14:24.399 --> 00:14:27.759
If in in that five minutes, I'm literally looking at the chain of thought.
00:14:27.919 --> 00:14:32.879
I'm making sure that it's calling the right tools, I'm making sure it's pulling in all the correct dynamic data.
00:14:32.960 --> 00:14:34.559
Because it's like, what else am I gonna do?
00:14:34.720 --> 00:14:41.759
Well, yeah, probably multitask with some uh some different agents and go in and check on my blue lights that you know require my input.
00:14:41.840 --> 00:14:46.879
But ultimately, I'm spending more time do it being the expert-driven loop, right?
00:14:47.120 --> 00:14:51.440
So I think that this is actually could be a big problem, right?
00:14:51.519 --> 00:14:59.200
When AI is almost too fast, even the people that are maybe trying to be responsible, uh, right, all of a sudden everyone knows.
00:14:59.279 --> 00:15:01.200
I mean, you know, in a year or so, everyone's gonna know.
00:15:01.360 --> 00:15:04.720
Oh, the AI is gonna be way more powerful and it's gonna be even faster.
00:15:04.960 --> 00:15:09.840
Whereas right now, if you're using AI in the right way, in theory, right, you're getting more done.
00:15:09.919 --> 00:15:15.039
But in you, it's almost like you're going slower, but way faster.
00:15:15.679 --> 00:15:21.200
You know, it's like you're waiting, but you're accomplishing more, right?
00:15:21.279 --> 00:15:31.679
So instead of working on how we would traditionally do it, you know, in pre-AI days, you know, me, one human, I would work on one project across these 20 little steps.
00:15:31.759 --> 00:15:38.960
Now, conversely, I'm working on 20 little projects, and my only one step is just checking in on what the agents are doing, right?
00:15:39.039 --> 00:15:54.000
But uh, you know, usually I stagger it in a way where I can um understand, um, kind of you know, prioritize um explainability and and you know, making sure the guardrails that I've set are being followed, making sure my agents aren't crashing, all of these things.
00:15:54.080 --> 00:15:55.519
But what happens when this is done?
00:15:55.759 --> 00:15:57.279
Am I still gonna go and do that?
00:15:57.440 --> 00:15:58.159
I hope so.
00:15:58.320 --> 00:16:07.360
But when things are done faster and when the models are smarter, won't my human nature just be to be like, well, I should go grow the business in a new way.
00:16:07.519 --> 00:16:09.679
I should go get more and more things done.
00:16:09.840 --> 00:16:10.480
Um, right.
00:16:10.639 --> 00:16:15.279
A fast just team just now needs more management.
00:16:15.919 --> 00:16:20.320
So directing agents is just middle management without the manager title, right?
00:16:20.399 --> 00:16:37.039
I do see the future um where most uh you know, most knowledge uh, you know, subject matter experts, most people, if if you're a knowledge worker, you sit in front of a computer um and you create value for your business, which is what many of us do.
00:16:37.200 --> 00:16:40.960
Um, I think people for the most part, you're gonna be like a middle manager, right?
00:16:41.120 --> 00:16:42.480
Of just agents.
00:16:43.360 --> 00:16:49.919
I like I I actually think that uh in theory, the middle management layer is going to get bigger and thicker.
00:16:50.000 --> 00:16:52.720
Um, I think there's gonna be fewer people at the entry level.
00:16:52.799 --> 00:16:57.600
I think there's gonna be fewer people in the, you know, the VP, C suite director level.
00:16:57.759 --> 00:17:04.160
I think for the most part, you know, your average knowledge worker, that middle uh middle management tier, right?
00:17:04.240 --> 00:17:12.160
When we think it's not really needed, well, I think true, you know, if we look at middle management today in an AI native world, it's not needed.
00:17:12.240 --> 00:17:15.759
But I think just everyone else is gonna become kind of that middle layer, right?
00:17:15.920 --> 00:17:18.559
People are gonna be managing more agents, not people.
00:17:18.720 --> 00:17:22.559
And yes, there's still gonna be, you know, humans reporting up the food chain.
00:17:22.720 --> 00:17:30.880
Um, but I think for the most part, the overwhelming majority of work is actually gonna get done in that middle management layer, where traditionally that's not where the work gets done, right?
00:17:31.039 --> 00:17:32.319
They're just managing people.
00:17:32.480 --> 00:17:35.599
Um, so it's I I think that's ultimately what's gonna happen.
00:17:35.839 --> 00:17:43.359
But I think it's normal in AI native companies for decision makers to just publish an agent to help to speed up productivity, right?
00:17:43.519 --> 00:17:59.759
You've seen these uh, you know, stories where literally, you know, a CEO or a VP or you know, someone that manages a big department at a huge company is just literally unloading their entire brain and decision-making process and all their documents and all of these things into an agent, right?
00:17:59.839 --> 00:18:09.119
Let's say it's it's you know, marketing Jane at a Fortune 500 company, you know, Jane just you know dumps her brain, her decision-making process, everything into an agent.
00:18:09.200 --> 00:18:14.240
And the you know, 500 people in marketing, well, they can just get answers straight from the Jane agent.
00:18:14.400 --> 00:18:15.680
They don't necessarily have to wait.
00:18:15.839 --> 00:18:19.519
So the work, um, it just becomes faster, but it becomes four verbs.
00:18:19.599 --> 00:18:26.559
I think you decide what matters, you delegate, you inspect the exceptions, and then you integrate the results, right?
00:18:26.720 --> 00:18:31.279
So, but those things can happen now like agents can in parallel.
00:18:31.440 --> 00:18:37.759
And you have to push that pattern to its limit, and that one person just directs more hours than exist, right?
00:18:37.839 --> 00:18:48.640
That's that's where I find myself and I found myself, you know, since uh kind of uh Claude Cowork um and Claude Code kind of kicked off this phase in late 2025, early 2026.
00:18:48.720 --> 00:18:53.039
And I think Codex um has really took in uh taken over the narrative ever since.
00:18:53.119 --> 00:18:59.039
But you know, essentially I'm directing more um hours than exist in a day, right?
00:18:59.119 --> 00:19:03.599
I'm literally getting done what used to take me even two years ago, right?
00:19:03.680 --> 00:19:11.839
Even in the early Chat GPT days, I'm getting uh about 30 hours of 2024 me work done every single day.
00:19:12.000 --> 00:19:17.599
So it's just you're directing more hours, more traffic than can fit on a traditional runway.
00:19:17.839 --> 00:19:23.359
Um, and you know, in uh OpenAI report came out with this, they showed how their team was using codex.
00:19:23.599 --> 00:19:32.720
And uh what I found interesting was uh OpenAI staff with a 99th percentile generated over 60 hours of agent turns every single day.
00:19:32.880 --> 00:19:35.039
So, yeah, that's kind of the anomaly, right?
00:19:35.119 --> 00:19:46.000
That's not the average because they're building this technology, but I think eventually that is going to become the default, and that is gonna force people, I think, to question how do you work, right?
00:19:46.160 --> 00:19:48.720
What is the value of my actual time?
00:19:49.039 --> 00:19:55.119
And how can we responsibly, you know, use these agents as they get even smarter and even faster.
00:19:55.279 --> 00:19:58.480
I actually think we're in a nice little groove where we are today, right?
00:19:58.559 --> 00:20:05.519
There's been all this, you know, talk uh in the last, you know, 36 hours or so of this concept of AI pacing, you know.
00:20:05.759 --> 00:20:19.359
Um uh essentially a lot of researchers, um particularly at OpenAI Anthropic, but also Google, Meta and other labs, you know, kind of sign this letter saying, hey, we support, you know, essentially calling it like pacing, right?
00:20:19.440 --> 00:20:23.519
But it's kind of just means like, hey, it's okay if we slow down.
00:20:23.680 --> 00:20:24.480
Will that happen?
00:20:24.720 --> 00:20:25.599
Probably not.
00:20:25.759 --> 00:20:26.240
Um, right.
00:20:26.400 --> 00:20:30.720
Is it the responsible thing for you know some sort of AI pacing to happen?
00:20:30.960 --> 00:20:31.680
Probably.
00:20:31.839 --> 00:20:36.240
Um, because the issue is the capability gap is growing.
00:20:36.400 --> 00:20:40.079
The model overhang is growing and it's not going to get any slower, right?
00:20:40.240 --> 00:20:50.720
Especially as uh, you know, over the past couple of months, we've gotten, you know, real results uh from the big labs talking about RSI recursive self-improvement, where the models are improving themselves, right?
00:20:50.880 --> 00:20:55.359
Yesterday, OpenAI literally just announced like an 18% improvement via RSI.
00:20:55.519 --> 00:21:01.920
They essentially said, Hey, we got, you know, we use GPT 5.6 to make itself better, and it's like 18% better.
00:21:02.000 --> 00:21:04.559
It's gonna see, you know, 18% cost savings now.
00:21:04.720 --> 00:21:07.759
Um, so this is where it's at.
00:21:07.839 --> 00:21:15.599
These models are gonna get better and better, faster and faster, but way at a much faster accelerated rate than humans can keep up with.
00:21:15.759 --> 00:21:19.839
Um, so yeah, there's this whole pacing thing that maybe we'll tackle at a different time.
00:21:20.079 --> 00:21:29.519
But the the the brutal reality is if you want your team to be able to keep up and get ahead, that's something that you have to tackle right now, head on.
00:21:29.599 --> 00:21:36.319
Uh, some other good stats that recently came out, a workday foundation study in May found that 86% of users felt more productive.
00:21:36.480 --> 00:21:42.799
But those same workers describe the conversations um that rarely moved past transactional work needs, right?
00:21:42.960 --> 00:21:44.240
And that's kind of where I feel right now.
00:21:44.319 --> 00:21:45.920
It's like, yeah, I'm getting way more done.
00:21:46.079 --> 00:21:48.400
The quant the content I hope is good.
00:21:48.559 --> 00:21:49.440
The quality is good, right?
00:21:49.519 --> 00:21:52.160
What I produce is much different than what most people produce.
00:21:52.319 --> 00:21:55.680
Um, but it's just kind of transactional, right?
00:21:55.759 --> 00:21:59.759
I've talked many times on the show about I'm not even.
00:22:00.319 --> 00:22:12.880
Because my you know, the throughput, the output is so insanely high for what I'm doing on a daily basis as an individual full-time human, I don't retain information like I used to because I think so much information now is transactional.
00:22:13.039 --> 00:22:18.720
And maybe that's something that we, you know, the future workforce needs to understand.
00:22:18.799 --> 00:22:26.480
Um, because I think, you know, have you ever thought about how you can learn a concept better if you write it down by hand or if you teach it to someone, right?
00:22:26.640 --> 00:22:30.160
I'm a big believer in you haven't really learned something until you teach, right?
00:22:30.240 --> 00:22:33.200
That's uh one of the reasons why I started a daily podcast.
00:22:33.279 --> 00:22:35.200
I'm like, I really want to learn this AI thing.
00:22:35.359 --> 00:22:39.440
I need to learn it well enough that I can teach at least one person, uh, right.
00:22:39.599 --> 00:22:41.680
Luckily, hopefully it's helped more than one person.
00:22:41.839 --> 00:22:43.839
Um, but I think the same is true.
00:22:44.000 --> 00:22:48.640
Um, because AI, it it does all that hard work for you, right?
00:22:48.720 --> 00:22:53.599
You no longer have to go onto 20 web pages and read it all and be like, ah, this doesn't make sense.
00:22:53.759 --> 00:22:54.799
This does make sense.
00:22:55.039 --> 00:22:56.079
I'm not sure about this.
00:22:56.160 --> 00:22:57.359
Let me look at a little more.
00:22:57.519 --> 00:23:03.440
Now you just put something into whatever AI system you're using, you get a personalized, customized output.
00:23:03.599 --> 00:23:07.759
And well, most people don't even take the time to truly read it and understand it.
00:23:07.920 --> 00:23:15.839
So, you know, there is this uh disconnection uh between you know the productivity and disconnection that's shared in one workday.
00:23:16.000 --> 00:23:19.279
And this study shows that that is the mechanism.
00:23:20.160 --> 00:23:23.680
Um, and I've actually, you know, touched on some of these other concepts.
00:23:23.839 --> 00:23:32.400
So, you know, if if you do want to, you know, dive in a little bit deeper on, you know, some of these uh downsides of using AI the right way.
00:23:32.480 --> 00:23:35.200
I actually had an episode, episode 757.
00:23:35.359 --> 00:23:37.279
Uh, that was part of the start here series.
00:23:37.440 --> 00:23:44.400
It was called The Seven Silent Sins of Doing AI Right, How to Spot and Overcome the Invisible AI work traps.
00:23:44.480 --> 00:23:52.319
And three that I wanted to share that are kind of relevant to what we're talking about today is number one, accidental de-skilling, which is kind of one of the things I just referenced there.
00:23:52.480 --> 00:23:53.680
That's where AI does the work.
00:23:53.839 --> 00:23:55.119
So you lose the skill.
00:23:55.200 --> 00:23:58.000
Uh, I talked about the agent bun sandwich.
00:23:58.079 --> 00:24:01.599
Um, and that's when agents start to hollow out your core expertise.
00:24:01.759 --> 00:24:05.680
And then the compression tax, where the AI speed overloads your brain.
00:24:05.839 --> 00:24:08.559
And that's where I think we might be getting into, right?
00:24:08.640 --> 00:24:11.759
With this um, you know, this cerebrus, right?
00:24:11.839 --> 00:24:15.680
Which might sound like a dorky niche thing, and you're like, okay, why do you keep talking about it?
00:24:15.839 --> 00:24:27.119
Because again, think if you are an AI power user now, and I said, hey, tomorrow, you or your company, your team can accomplish 20x what you accomplished today.
00:24:27.279 --> 00:24:34.240
If you are an agent-pilled team, and I'm like, you can accomplish 20 times more today, that's some compression tax happening right there.
00:24:34.400 --> 00:24:34.640
All right.
00:24:34.720 --> 00:24:38.160
So yeah, make sure uh go check out episode 757.
00:24:38.720 --> 00:24:40.880
So let's be fair to AI, though.
00:24:41.039 --> 00:24:43.359
Some of this stuff, we don't need it, right?
00:24:43.440 --> 00:24:47.680
How many times have you gone through a three-hour meeting and you're like, that could have been a single bullet point?
00:24:47.839 --> 00:24:49.839
That could have been a single Slack message, right?
00:24:50.000 --> 00:24:55.680
Or waiting two days for a basic answer is not, you know, culture or mentorship.
00:24:55.759 --> 00:25:01.680
It's just maybe the signs of an antiquated business hierarchy that shouldn't have existed in the first place, right?
00:25:01.759 --> 00:25:09.039
I think most companies um didn't quickly redesign how they worked um for an AI native workplace.
00:25:09.279 --> 00:25:17.920
So, you know, uh you get to then decide um which collaboration is intentional.
00:25:18.319 --> 00:25:25.119
Um, you know, human to human, you get to choose, um, kind of pick your spots where that happens if you are AI native.
00:25:25.599 --> 00:25:29.039
Okay, so as we wrap up here, I want to try to answer this question.
00:25:29.200 --> 00:25:41.839
How do you strike the balance then of you know the reality of we're gonna be managing more agents, which might lead to less human connection and probably an increased expectation of productivity?
00:25:41.920 --> 00:25:44.799
How do all of those three things uh build together?
00:25:44.960 --> 00:25:47.519
And I know this isn't for 100% of organizations, right?
00:25:47.599 --> 00:25:54.000
I I understand that, but I do think that more organizations, more teams will be shifting toward this than the opposite.
00:25:54.240 --> 00:25:55.759
So, how do you meet that head on?
00:25:55.839 --> 00:26:03.680
Well, I think that winning companies are gonna intentionally automate the manual work, but keep and prioritize human judgment.
00:26:03.920 --> 00:26:07.119
So I think generic human in the loop will always fail.
00:26:07.279 --> 00:26:09.359
So you have to name one expert owner.
00:26:09.599 --> 00:26:21.519
I always talk about expert-driven loops, and expert-driven loops actually require instead of one passive human generalist, it requires active expert human collaboration and communication.
00:26:21.680 --> 00:26:23.039
So I think that's one thing right there.
00:26:23.200 --> 00:26:30.400
But also, I think winning teams um know which friction to remove and which human handoffs to protect and maybe elevate.
00:26:30.640 --> 00:26:32.880
So then how do we strike the balance then?
00:26:33.680 --> 00:26:41.839
Well, I think, yeah, as as my slides are losing, losing my place here.
00:26:42.079 --> 00:26:45.200
Uh sorry, yeah, all all screwed up.
00:26:46.000 --> 00:26:46.640
There we go.
00:26:46.799 --> 00:26:48.319
My slides got all out of order.
00:26:48.480 --> 00:26:53.119
So um, here's the three human handoffs that I think you need to protect.
00:26:53.279 --> 00:27:00.240
So, number one, you need to protect the judgment by naming the human who owns consequences when the answer is wrong.
00:27:00.480 --> 00:27:00.799
Okay.
00:27:01.119 --> 00:27:09.039
Um, number two, you need to protect learning by making people uh explain decisions instead of blindly approving agent output.
00:27:09.119 --> 00:27:24.720
So that's kind of the difference between, like I said, passive, lazy, human in the loop, which will always fail, and the correct way to do it, which is expert-driven loops, which is experts um communicating uh with each other, but also expecting, um inspecting agent outputs.
00:27:24.799 --> 00:27:27.680
And then last, you need to protect your belonging, right?
00:27:27.759 --> 00:27:34.160
I think so much of this, um, so much of what we do in our work, it's about having this sense of belonging.
00:27:34.319 --> 00:27:44.480
It's about, you know, some people, for better or worse, wrap their um, you know, self in um their own meaning is who they are in their job, right?
00:27:44.559 --> 00:27:52.160
But I think you can still protect your belonging by giving coaching, recognition, and unstructured conversation a scheduled place to happen.
00:27:52.400 --> 00:27:52.880
All right.
00:27:53.039 --> 00:27:57.119
So, what's my big takeaway here as we get faster agents that are smarter?
00:27:57.359 --> 00:28:04.640
And maybe we're just managing dozens or hundreds of agents and talking to our human coworkers less, but producing more.
00:28:04.799 --> 00:28:07.440
Well, I think three big takeaways for me.
00:28:07.599 --> 00:28:10.960
Number one, everyone needs to increase their ambition, right?
00:28:11.039 --> 00:28:17.599
Not just saying more prompts, more outputs, but being more intentional with what you can accomplish, right?
00:28:17.759 --> 00:28:22.960
Maybe not just doing 20x of what you were doing yesterday because you can do 20x today.
00:28:23.119 --> 00:28:28.240
Maybe instead it's thinking higher, it's thinking harder, it's just being more ambitious.
00:28:28.400 --> 00:28:32.000
Number two, you always have to unmarry your domain expertise.
00:28:32.240 --> 00:28:38.319
That is the uh, you know, the building blocks of what I've been talking about now for three and a half years, unlearning.
00:28:38.400 --> 00:28:43.359
Um, you have to unmarry your domain expertise because it doesn't really matter necessarily anymore.
00:28:43.519 --> 00:28:49.039
Your domain expertise is really just about that, you know, the human agent bun sandwich, right?
00:28:49.119 --> 00:28:55.519
It's it's making sure that you give it the right context and check it in the right way, but you're not going to be doing the 95% in the middle anymore.
00:28:55.599 --> 00:28:59.519
So you have to unmarry your domain expertise as agents get faster and smarter.
00:28:59.680 --> 00:29:05.200
And then last but not least, you have to be prepared to rebuild monthly instead of yearly.
00:29:05.359 --> 00:29:22.000
Um, I think in the pre-agentic phase of 2024 or early 2025, you know, I was still recommending that people revisit their AI strategy roadmaps quarterly, but I would say even if you did it yearly, you know, in 2024, you weren't gonna be left behind.
00:29:22.079 --> 00:29:22.880
Now you will.
00:29:23.039 --> 00:29:34.720
You have to be prepared to literally uh unlearn and rebuild how your company operates every single month, which sounds like an absolutely asinine concept.
00:29:34.880 --> 00:29:38.000
Because before you only had to do it every five years, right?
00:29:38.160 --> 00:29:40.079
Uh you know, or maybe every 10 years.
00:29:40.160 --> 00:29:52.880
And then, you know, as the internet and you know, web two and social media and all these other things uh started to gain in popularity and how people make decisions and all that, maybe, you know, advertising, marketing, communications, you know, those things changes.
00:29:52.960 --> 00:30:08.000
And so, you know, with digital transformation, I think that shortened um that lifespan of you know how your company works, kind of your uh your company's IP, not just your SOPs, but literally how your company works and runs and thinks and makes decisions.
00:30:08.160 --> 00:30:09.519
You can't do it anymore.
00:30:09.759 --> 00:30:25.279
As crazy as it sounds, I think probably a majority of the time for uh successful companies moving forward are going to spend as much time uh unlearning and rebuilding processes as they are actually doing work the old-fashioned way.
00:30:25.440 --> 00:30:25.759
All right.
00:30:25.920 --> 00:30:28.079
So I hope this episode was helpful.
00:30:28.240 --> 00:30:36.079
Just more of a kind of a zoomed out look of saying, I don't necessarily know what the heck is going on or what the heck is happening next.
00:30:36.240 --> 00:30:54.319
But if you look at the writing on the wall, you have to see and you have to understand agents are getting smarter, agents are getting faster, and we, as people who are AI leaders in our respective organizations, well, we're gonna be working with way more agents and probably fewer humans in many cases.
00:30:54.480 --> 00:30:57.279
Um, so I hope this episode was helpful.
00:30:57.359 --> 00:30:58.240
If so, let me know.
00:30:58.319 --> 00:30:59.680
Uh, please subscribe to the podcast.
00:30:59.759 --> 00:31:06.319
If you want to hear more or less of shows like this, uh, do let me know by signing up for the newsletter at your everydayai.com.
00:31:06.400 --> 00:31:07.359
So thank you for tuning in.
00:31:07.519 --> 00:31:10.240
Hope to see you back tomorrow and every day for more everyday AI.
00:31:10.319 --> 00:31:11.359
Thanks, y'all.
