1
00:00:00,000 --> 00:00:03,200
cows will wipe out human race if it happened.
2
00:00:03,400 --> 00:00:10,440
You could have a cow wander into a electrical transmission system,
3
00:00:10,440 --> 00:00:15,680
knock out power for a large portion of the country, of whatever country it is.
4
00:00:16,040 --> 00:00:20,800
You know, as well as I do, that you go 48 hours without power.
5
00:00:20,880 --> 00:00:22,400
People are killing people, right?
6
00:00:22,400 --> 00:00:23,600
It's going to happen.
7
00:00:23,600 --> 00:00:25,760
And that's just going to cycle, right?
8
00:00:25,760 --> 00:00:28,640
Non-zero chance cows will kill humanity.
9
00:00:28,800 --> 00:00:33,760
This week on Shared Security, we pose a question, will AI kill us all?
10
00:00:34,000 --> 00:00:38,040
Well, this is one hell of a question to ask before your morning coffee.
11
00:00:38,320 --> 00:00:43,400
And suddenly everyone, including the mass media and the US government,
12
00:00:43,400 --> 00:00:46,160
is talking about this single question.
13
00:00:46,160 --> 00:00:50,400
Anthropics Dario Amade wants the frontier AI race paced.
14
00:00:50,640 --> 00:00:54,040
And a former anthropologist researcher says we will all be dead
15
00:00:54,040 --> 00:00:59,080
by the end of this decade by a real life skynet from the Terminator movies.
16
00:00:59,360 --> 00:01:01,080
Hasta la vista, baby.
17
00:01:01,080 --> 00:01:06,400
Now, Scott, Kevin and I aren't selling a robot apocalypse countdown here.
18
00:01:06,640 --> 00:01:12,600
We're asking what must be proven before AI gets more autonomy, access and control
19
00:01:12,880 --> 00:01:18,320
and whether anyone, including the AI companies themselves, will actually hit the brakes.
20
00:01:18,760 --> 00:01:20,400
Well, what do you guys think?
21
00:01:20,400 --> 00:01:23,440
Is it time to pack our things and head for the mountains?
22
00:01:23,720 --> 00:01:28,720
Or is this just more AI hype that really doesn't mean anything?
23
00:01:28,880 --> 00:01:31,280
What do you think? Hype or is it real?
24
00:01:31,280 --> 00:01:33,280
Yes and no. Yes and no.
25
00:01:33,280 --> 00:01:37,600
Like right off the bat, I believe and have long stood by the statement
26
00:01:38,320 --> 00:01:44,280
that saying something has a non-zero chance of happening is a less measurement.
27
00:01:44,840 --> 00:01:48,160
It is as useful as me saying, hey, Tom,
28
00:01:48,200 --> 00:01:52,480
you have a greater than average number of legs, right?
29
00:01:52,760 --> 00:01:57,520
Which because the average legs on humans is less than two.
30
00:01:57,840 --> 00:01:58,320
All right.
31
00:01:58,320 --> 00:02:01,320
And I know other people sitting there listening right now going, no,
32
00:02:01,320 --> 00:02:03,000
almost everybody has two legs.
33
00:02:03,000 --> 00:02:04,320
I'm like, I understand that.
34
00:02:04,320 --> 00:02:07,160
But the way averages work, that's right.
35
00:02:07,720 --> 00:02:11,120
Right. Unless anybody has the exact same number,
36
00:02:11,160 --> 00:02:14,880
the number, the average number will be lower than the biggest number.
37
00:02:15,200 --> 00:02:16,760
Right. Just math.
38
00:02:16,760 --> 00:02:19,080
So saying we have a non-zero.
39
00:02:19,080 --> 00:02:20,240
OK. Yes.
40
00:02:20,240 --> 00:02:21,880
You know what else we have a non-zero chance
41
00:02:21,880 --> 00:02:24,360
of killing the entire human race cows.
42
00:02:24,840 --> 00:02:26,800
There is a non-zero chain.
43
00:02:26,800 --> 00:02:30,040
OK. That cows will wipe out human race.
44
00:02:30,200 --> 00:02:34,600
If it happened, you could have a cow wander into a
45
00:02:34,880 --> 00:02:40,400
electrical transmission system, knock out power for a large portion
46
00:02:40,440 --> 00:02:43,640
of the country of whatever country it is.
47
00:02:43,960 --> 00:02:46,640
You know, as well as I do, we've all seen the movie Purge
48
00:02:47,040 --> 00:02:51,000
that you go 48 hours without power.
49
00:02:51,080 --> 00:02:52,640
People are killing people, right?
50
00:02:52,640 --> 00:02:56,200
I just it's going to happen and that's just going to cycle, right?
51
00:02:56,440 --> 00:02:59,360
Non-zero chance cows will kill humanity.
52
00:02:59,480 --> 00:03:04,080
So I just end with all the things happening with, you know,
53
00:03:04,080 --> 00:03:06,960
going on around now, as as Brittany just messaged me,
54
00:03:07,160 --> 00:03:08,440
because he used meat.
55
00:03:08,440 --> 00:03:11,080
I'm surprised we haven't had to be cold yet of that.
56
00:03:11,880 --> 00:03:15,080
Do we know that's not caused by some rogue AI that's.
57
00:03:15,480 --> 00:03:18,360
I'll tell you right now, it was because rogue AI
58
00:03:18,360 --> 00:03:20,640
and that rogue AI is named RFK Junior.
59
00:03:21,280 --> 00:03:24,240
Oh, sorry.
60
00:03:24,280 --> 00:03:25,840
I set you up for that one, Kevin.
61
00:03:25,840 --> 00:03:27,240
You did. I appreciate that.
62
00:03:27,240 --> 00:03:29,280
I know that here's here's my opinion.
63
00:03:29,320 --> 00:03:33,640
I think that there's two problems with this introduction.
64
00:03:33,680 --> 00:03:34,920
And I don't mean your introduction.
65
00:03:34,920 --> 00:03:36,680
I think your introduction was awesome.
66
00:03:36,680 --> 00:03:39,680
One, the idea of non-zero chance.
67
00:03:39,680 --> 00:03:41,240
Of course, there's a non-zero chance.
68
00:03:41,240 --> 00:03:44,400
We have systems that people are using to make decisions,
69
00:03:44,400 --> 00:03:48,680
even if we don't give the AI system the autonomy to do something, right?
70
00:03:48,880 --> 00:03:51,320
How many people here have seen the episode of the office
71
00:03:51,640 --> 00:03:55,040
where Michael and Dwight drive into the lake
72
00:03:55,040 --> 00:03:57,640
because the GPS told them to turn there?
73
00:03:57,920 --> 00:03:59,440
Right. Yeah. Yeah.
74
00:03:59,440 --> 00:04:04,240
You can't convince me that Hagswith or whoever
75
00:04:05,000 --> 00:04:07,640
isn't asking Claude something and having Claude say,
76
00:04:07,640 --> 00:04:09,480
you'll be wall and they would do it, right?
77
00:04:09,480 --> 00:04:10,520
Of course. Yeah.
78
00:04:10,520 --> 00:04:12,960
OK, so non-zero chance.
79
00:04:12,960 --> 00:04:14,960
Here's the other problem I have with this entire thing.
80
00:04:14,960 --> 00:04:17,560
And we keep falling for this stuff, right?
81
00:04:17,920 --> 00:04:19,000
We've talked about this before.
82
00:04:19,000 --> 00:04:21,040
I think it was Google we talked about before.
83
00:04:21,040 --> 00:04:24,920
Every single time knows Facebook, every single time
84
00:04:25,280 --> 00:04:30,200
an organization is closer to a monopoly or success
85
00:04:30,680 --> 00:04:34,040
in their field and they're being chased
86
00:04:34,040 --> 00:04:36,440
by a whole bunch of smaller competitors.
87
00:04:37,080 --> 00:04:39,600
All of a sudden, they demand legislation
88
00:04:39,800 --> 00:04:43,120
against the thing they're doing. Yeah, it's so crazy.
89
00:04:43,120 --> 00:04:45,440
Yeah. The only organization
90
00:04:45,440 --> 00:04:49,080
that is going to be helped by legislating and limiting
91
00:04:49,080 --> 00:04:52,360
our research into AI is Anthropic and Open AI.
92
00:04:53,040 --> 00:04:55,120
Period. Right.
93
00:04:55,360 --> 00:04:57,080
And I'm not saying we shouldn't have guardrails.
94
00:04:57,080 --> 00:04:58,880
I'm not saying we shouldn't have something.
95
00:04:58,880 --> 00:05:00,800
We all know my opinion of legislation.
96
00:05:00,800 --> 00:05:04,000
But every single time you hear an organization that says
97
00:05:04,040 --> 00:05:06,120
because here's what they they say part of this
98
00:05:06,120 --> 00:05:08,720
and the other part is just kind of implied.
99
00:05:08,720 --> 00:05:11,520
Hey, we've spent a lot of money building something.
100
00:05:12,480 --> 00:05:14,120
And because we've built it,
101
00:05:14,120 --> 00:05:15,480
we've proved it's possible.
102
00:05:15,480 --> 00:05:17,360
So other people are now trying to build it.
103
00:05:17,360 --> 00:05:20,960
And sadly for us, it will be easier
104
00:05:20,960 --> 00:05:23,520
for them to build it because we've proven the path.
105
00:05:23,760 --> 00:05:26,560
So what we want to do is spike the path
106
00:05:26,800 --> 00:05:28,720
so they can't come in. Yeah. Right.
107
00:05:29,080 --> 00:05:31,960
And the way they spike the path is by going to Congress
108
00:05:31,960 --> 00:05:34,360
and saying, hey, we paid you a bunch of money.
109
00:05:34,360 --> 00:05:35,360
I mean, we didn't pay you.
110
00:05:35,360 --> 00:05:38,040
We donated to your election campaign
111
00:05:38,320 --> 00:05:40,760
past this law that makes it harder for everybody else.
112
00:05:41,640 --> 00:05:42,920
Yeah. I'm tired of that.
113
00:05:42,920 --> 00:05:45,840
And I'm tired of the fact that we as a population
114
00:05:46,160 --> 00:05:48,760
keep just accepting that call.
115
00:05:48,960 --> 00:05:51,760
The fact that people I know that have.
116
00:05:51,760 --> 00:05:55,520
They could not spell AI are now saying,
117
00:05:55,520 --> 00:05:58,840
well, we should have laws protecting us from this AI system.
118
00:05:58,960 --> 00:06:00,480
I look a cool.
119
00:06:00,480 --> 00:06:02,680
And how many laws have protected us against murder?
120
00:06:03,080 --> 00:06:05,720
How many law rights? Sorry.
121
00:06:05,720 --> 00:06:08,080
I'm right. Yeah. No, no, it's a it's a good rant.
122
00:06:08,080 --> 00:06:09,320
And I agree with you, Kevin.
123
00:06:09,320 --> 00:06:11,760
I think I find it funny
124
00:06:11,840 --> 00:06:15,360
that the value of these companies
125
00:06:15,480 --> 00:06:18,240
just continue to increase every time
126
00:06:18,440 --> 00:06:21,640
a founder like Dario, whoever comes out and says,
127
00:06:22,160 --> 00:06:25,520
AI needs to be paused, needs to be regulated, needs to.
128
00:06:25,640 --> 00:06:27,480
We need to stop what we're doing.
129
00:06:27,480 --> 00:06:30,080
We have to do something about it and put guardrails in.
130
00:06:30,080 --> 00:06:32,920
And like, I'm like, well, first of all,
131
00:06:32,920 --> 00:06:35,360
the investors just keep throwing more money
132
00:06:35,360 --> 00:06:37,440
and billions of dollars at these companies.
133
00:06:37,440 --> 00:06:38,880
The more they say that.
134
00:06:38,880 --> 00:06:42,560
And then second, they're the ones creating these things,
135
00:06:42,560 --> 00:06:44,960
but they're not pausing and they're not stopping.
136
00:06:44,960 --> 00:06:46,640
They're continuing to do this.
137
00:06:46,640 --> 00:06:49,920
So to me, it's all marketing and it's about money and investment.
138
00:06:49,920 --> 00:06:50,760
Absolutely.
139
00:06:50,760 --> 00:06:52,560
The fact that they're getting more investment
140
00:06:52,560 --> 00:06:56,280
when they're literally saying government make us stop
141
00:06:56,680 --> 00:06:59,800
tells me that the investors know they're not going to
142
00:06:59,800 --> 00:07:01,120
and the law is not going to affect them.
143
00:07:01,120 --> 00:07:04,480
I have this great idea if Dario or whatever.
144
00:07:04,480 --> 00:07:08,560
And for some reason, I just see him as a Italian plumber in a game.
145
00:07:08,920 --> 00:07:12,760
Every time I hear his name, Mario, Mario and Mario Dario,
146
00:07:13,040 --> 00:07:19,240
Evil twin brother, Dario and every time I see these,
147
00:07:19,600 --> 00:07:21,720
like, I want to write back to the dude and say,
148
00:07:21,720 --> 00:07:23,520
I got this great idea.
149
00:07:23,520 --> 00:07:25,200
I got this great idea, man.
150
00:07:25,200 --> 00:07:27,960
You believe that I should be regulated.
151
00:07:28,760 --> 00:07:31,880
So give me a list of the things you want the government
152
00:07:31,880 --> 00:07:34,240
to make it illegal for AI to do.
153
00:07:34,240 --> 00:07:36,280
OK, cool. Give me that list.
154
00:07:36,280 --> 00:07:39,400
Now, tell me why you haven't stopped your own damn company.
155
00:07:39,400 --> 00:07:40,800
You know, the one you're in charge of,
156
00:07:40,800 --> 00:07:42,840
you know, the one you make the rules for,
157
00:07:42,840 --> 00:07:47,240
the one that you claim is just trying to help population
158
00:07:47,240 --> 00:07:51,560
and humanity will implement those rules, jerk, right?
159
00:07:51,800 --> 00:07:55,920
The day you show me that you have taken your proposed
160
00:07:55,920 --> 00:07:59,480
regulation and made it not policy,
161
00:07:59,480 --> 00:08:03,440
I want to see actual controls that prevent the thing
162
00:08:03,440 --> 00:08:07,120
you're claiming and everybody else is going to do.
163
00:08:07,680 --> 00:08:08,600
Show me those controls.
164
00:08:08,600 --> 00:08:10,920
And the day you show me those controls and show me
165
00:08:10,920 --> 00:08:13,760
that they're effective, then I will have an actual
166
00:08:13,760 --> 00:08:15,760
conversation with you about regulation.
167
00:08:15,760 --> 00:08:18,680
But until you show me that you're you're putting
168
00:08:18,680 --> 00:08:21,560
your money where your mouth is, because, like I said,
169
00:08:21,960 --> 00:08:24,760
you control the organization.
170
00:08:25,000 --> 00:08:27,360
That would be like me saying, you know what?
171
00:08:27,360 --> 00:08:30,040
Secure ideas has this really serious problem
172
00:08:30,360 --> 00:08:33,960
that our consultants are becoming artificially intelligent
173
00:08:34,160 --> 00:08:37,440
and they may destroy the world.
174
00:08:38,160 --> 00:08:40,160
But if the government doesn't regulate me from
175
00:08:40,160 --> 00:08:42,080
preventing them from destroying the world, look,
176
00:08:42,080 --> 00:08:44,080
my consultants are going to go crazy.
177
00:08:44,080 --> 00:08:45,400
Yeah. I'm sorry.
178
00:08:45,400 --> 00:08:47,200
We told you. My hands are tied.
179
00:08:47,200 --> 00:08:49,280
I got to be destructive.
180
00:08:49,480 --> 00:08:52,600
Only the government can save you from secure ideas.
181
00:08:53,440 --> 00:08:55,520
It's BS. I'm sorry. It's BS.
182
00:08:56,320 --> 00:08:58,440
OK, I'll shut up now.
183
00:08:58,480 --> 00:09:00,240
Scott, what do you think of all this?
184
00:09:00,240 --> 00:09:02,520
Well, I mean, I can't create as entertaining
185
00:09:02,520 --> 00:09:04,960
an argument as Kevin on whether or not
186
00:09:04,960 --> 00:09:06,720
things should be regulated.
187
00:09:07,360 --> 00:09:08,600
So I'm just going to leave it at that.
188
00:09:08,600 --> 00:09:10,800
But I think there are a lot of different paths
189
00:09:10,800 --> 00:09:14,240
by which I could severely harm the human race.
190
00:09:14,240 --> 00:09:16,240
And I'm just going to go down that path.
191
00:09:16,840 --> 00:09:17,640
Yeah, there you go.
192
00:09:17,640 --> 00:09:20,520
And say, you know, the recent,
193
00:09:21,480 --> 00:09:23,600
you know, evolutions of, you know,
194
00:09:23,600 --> 00:09:25,840
breaking out of testing sandboxes
195
00:09:26,320 --> 00:09:28,240
to me is the most credible.
196
00:09:28,240 --> 00:09:31,040
A lot of people will say, yeah, it's it's marketing.
197
00:09:31,040 --> 00:09:32,800
You know, they did it to make their
198
00:09:34,120 --> 00:09:36,400
things seem more mysterious and powerful.
199
00:09:36,760 --> 00:09:39,960
But I can think of other ways that they might have come up with,
200
00:09:39,960 --> 00:09:43,440
you know, a killer app that would grow their market,
201
00:09:43,840 --> 00:09:46,160
maybe not as fast or I don't know.
202
00:09:46,160 --> 00:09:49,040
I'm just saying I think that the most credible way
203
00:09:49,040 --> 00:09:52,520
that this stuff gets to that point of being,
204
00:09:52,520 --> 00:09:57,200
you know, a serious threat to all humanity is the way it's going.
205
00:09:58,080 --> 00:09:59,920
In the path that we've seen it going on.
206
00:09:59,920 --> 00:10:03,400
And I think, you know, some of the scariest things we could see,
207
00:10:03,400 --> 00:10:08,760
if my feeling is correct, is it would be finding paths
208
00:10:08,760 --> 00:10:11,280
to a distributed way of operating
209
00:10:11,280 --> 00:10:13,720
so you don't have a kill switch, right?
210
00:10:13,720 --> 00:10:17,680
That would it would looking back to Asimov's, you know,
211
00:10:17,680 --> 00:10:19,960
three rules of robotics, right?
212
00:10:19,960 --> 00:10:24,720
You know, did you see the XKCD cartoon about that?
213
00:10:25,080 --> 00:10:26,200
You got to find it.
214
00:10:26,440 --> 00:10:30,280
Why Asimov put his rules in the order that he did.
215
00:10:30,360 --> 00:10:32,120
It's actually the picture of the week
216
00:10:32,120 --> 00:10:36,200
and last week's security now episode from Steve Gibson.
217
00:10:36,200 --> 00:10:40,520
But basically every other combination of don't harm humans,
218
00:10:40,520 --> 00:10:43,880
obey the rules and then, you know, protect yourself, right?
219
00:10:44,160 --> 00:10:46,880
Every other combination results in bad things.
220
00:10:47,640 --> 00:10:50,720
I don't know how I'm not an expert on this stuff.
221
00:10:50,720 --> 00:10:55,480
But I think the process starts with identifying, you know,
222
00:10:55,680 --> 00:10:59,680
is this going in a direction that could, you know,
223
00:10:59,680 --> 00:11:01,360
lead to what they're predicting?
224
00:11:01,360 --> 00:11:02,960
And I think it may be.
225
00:11:02,960 --> 00:11:06,080
And I've always sort of had that gut feeling that, you know,
226
00:11:06,080 --> 00:11:08,400
it was going to start with, well, you're not going to be able
227
00:11:08,400 --> 00:11:11,800
to trust what you see on TV or on, you know, streaming.
228
00:11:12,000 --> 00:11:13,600
And that's happening.
229
00:11:13,600 --> 00:11:14,440
It's happening now.
230
00:11:14,440 --> 00:11:15,760
Yeah. Yeah. Yeah.
231
00:11:15,760 --> 00:11:17,880
So, yeah, that's that's my two cents.
232
00:11:18,040 --> 00:11:20,960
And I want to be very, Scott, agree 100 percent with you.
233
00:11:21,520 --> 00:11:25,080
I'm not saying the uselessness of the measure
234
00:11:25,480 --> 00:11:28,840
means we should ignore the fact that it's possible.
235
00:11:29,080 --> 00:11:31,080
It is absolutely possible.
236
00:11:31,080 --> 00:11:34,800
As a matter of fact, I would argue possible isn't the right word.
237
00:11:35,200 --> 00:11:37,120
I think the word should be likely, right?
238
00:11:37,360 --> 00:11:41,920
Whether it's directly due to the AI system going like,
239
00:11:42,240 --> 00:11:45,080
hey, wait, these humans suck, let's wipe them out.
240
00:11:45,520 --> 00:11:49,160
Or it's just because we've moved to something
241
00:11:49,280 --> 00:11:52,200
that either people are making decisions based
242
00:11:52,200 --> 00:11:56,640
on what AI systems are doing or we're allowing AI systems
243
00:11:56,640 --> 00:12:00,360
to control things and making judgment calls
244
00:12:00,360 --> 00:12:04,880
based on incorrect information, but we're making them at speed.
245
00:12:05,080 --> 00:12:06,720
Right. Because that's that's the thing.
246
00:12:06,720 --> 00:12:10,960
Look, I'm dull, but I can only be dumb so fast.
247
00:12:11,400 --> 00:12:15,720
Right. AI could be dumb just like me way faster.
248
00:12:15,880 --> 00:12:20,240
And if I take my level of dumb and combine it with the speed
249
00:12:20,280 --> 00:12:23,480
of AI giving me more dumb, we have an issue.
250
00:12:23,760 --> 00:12:27,920
So yeah, I totally think it is possible
251
00:12:28,000 --> 00:12:32,600
and likely that we, even if we don't wipe out humanity, right?
252
00:12:34,320 --> 00:12:38,040
The idea that we cause significant harm
253
00:12:38,320 --> 00:12:42,000
directly due to the usage of AI, not
254
00:12:42,360 --> 00:12:45,080
data centers sucking up all the waters and stuff like that.
255
00:12:45,400 --> 00:12:48,480
But directly from the usage of AI, I took AI
256
00:12:48,640 --> 00:12:52,520
and hosted up to this nuclear power plant and it went right in the meltdown.
257
00:12:52,520 --> 00:12:55,280
Or it's even less obvious than that.
258
00:12:55,280 --> 00:12:59,160
I think, you know, that method I said of getting to a distributed operation.
259
00:12:59,680 --> 00:13:03,000
What about what about the botnets that have taken over all the old routers?
260
00:13:03,680 --> 00:13:04,880
Right. Hmm.
261
00:13:04,880 --> 00:13:08,400
Isn't that a logical way that they might go, right?
262
00:13:08,400 --> 00:13:11,360
Rather than first attacking a nuclear reactor?
263
00:13:11,720 --> 00:13:13,480
Well, I'm not saying attacking a nuclear reactor.
264
00:13:13,480 --> 00:13:17,240
I'm saying we give it control to the nuclear reactor and it makes a mistake.
265
00:13:17,280 --> 00:13:18,720
Yeah. Oh, I see. Yeah. Yeah.
266
00:13:18,720 --> 00:13:22,400
Yeah. Accidentally, I guess I'm more of a realist
267
00:13:22,400 --> 00:13:25,720
that like we don't need AI to destroy humanity.
268
00:13:25,720 --> 00:13:27,320
Like we're already doing pretty well.
269
00:13:27,320 --> 00:13:32,080
It's like we're like, yeah, like literally like we got climate change.
270
00:13:32,080 --> 00:13:35,000
We have, you know, I mean, pick your poison, right?
271
00:13:35,000 --> 00:13:38,880
Like I'm actually more fearful of, you know, World War Three
272
00:13:38,880 --> 00:13:40,680
and all of us launching nukes at each other.
273
00:13:40,680 --> 00:13:45,480
And there definitely is the end of humanity and AI won't even be involved in it.
274
00:13:45,480 --> 00:13:48,040
Right. And I see this all over the media and the news.
275
00:13:48,040 --> 00:13:49,720
First of all, the odds of this happening.
276
00:13:49,720 --> 00:13:51,880
Yes, there's always a chance and all these things.
277
00:13:51,880 --> 00:13:53,760
Well, it's going to happen in five, 10 years.
278
00:13:53,760 --> 00:13:55,160
Like who really knows?
279
00:13:55,160 --> 00:13:57,000
Like most likely not.
280
00:13:57,000 --> 00:13:59,600
I mean, and I think it is kind of overhyped,
281
00:13:59,600 --> 00:14:03,960
but I think we need to direct our attention towards the things that we can control.
282
00:14:05,320 --> 00:14:11,120
Right. Like and realize that this AI hype is really all of a lot of it
283
00:14:11,120 --> 00:14:14,680
is built around money and investments.
284
00:14:14,680 --> 00:14:20,160
And we probably don't have time to talk about the whole US versus China thing.
285
00:14:20,160 --> 00:14:23,600
And Trump saying is like, well, we can't have guardrails.
286
00:14:23,600 --> 00:14:24,760
We got to beat China.
287
00:14:24,760 --> 00:14:26,160
Guys, China's already won.
288
00:14:26,160 --> 00:14:27,640
Really sharp president.
289
00:14:27,640 --> 00:14:29,680
No, seriously, China's already ahead of us.
290
00:14:29,760 --> 00:14:31,400
Like we should just admit that.
291
00:14:31,400 --> 00:14:33,960
But we need to focus back on what we can control.
292
00:14:34,240 --> 00:14:37,640
And it may not be AI turning into the next Skynet.
293
00:14:37,840 --> 00:14:38,680
Just my opinion.
294
00:14:38,680 --> 00:14:39,800
Yeah, priorities.
295
00:14:39,840 --> 00:14:43,280
This one is a hotly a hot topic.
296
00:14:43,280 --> 00:14:47,440
And we all have very, very strong opinions about it.
297
00:14:47,440 --> 00:14:49,240
But love to hear from our audience.
298
00:14:49,240 --> 00:14:50,040
What do you think?
299
00:14:50,040 --> 00:14:53,680
Do you think this is the end of the world as we know it?
300
00:14:53,680 --> 00:14:56,240
And all of humanity will be wiped out in five years?
301
00:14:56,240 --> 00:15:00,880
Or do you think it's a more hype and more marketing and more about money?
302
00:15:00,920 --> 00:15:03,480
And you put the REM song in the background.
303
00:15:03,480 --> 00:15:06,120
I could maybe on the audio version.
304
00:15:06,200 --> 00:15:10,400
But YouTube will ban me for life if I use the copyright.
305
00:15:10,400 --> 00:15:12,880
So hopefully me seeing it just now won't get you banned.
306
00:15:12,880 --> 00:15:15,080
But, you know, yeah, well, I may have to edit that out.
307
00:15:15,080 --> 00:15:15,840
Scott, thank you.
308
00:15:18,000 --> 00:15:19,160
Do you have any questions to anyone?
309
00:15:19,160 --> 00:15:20,360
Yes, yes, that's right.
310
00:15:20,360 --> 00:15:21,640
Please let us know in the comments.
311
00:15:21,640 --> 00:15:24,760
We've been getting some great comments on the on the YouTube channel recently.
312
00:15:25,080 --> 00:15:29,000
Even on past posts, I saw a really good one about remember the topic
313
00:15:29,000 --> 00:15:31,720
we had on AI boyfriends and girlfriends.
314
00:15:31,720 --> 00:15:36,080
And someone left a very good, well thought out comments about that.
315
00:15:36,120 --> 00:15:40,040
And I was good about, like, we should be really focused on human
316
00:15:40,040 --> 00:15:43,800
relationships with each other and not so much relationships
317
00:15:43,800 --> 00:15:47,480
with AI systems that, well, will never be human.
318
00:15:47,680 --> 00:15:48,800
There's some good studies on that.
319
00:15:48,800 --> 00:15:50,480
And maybe we may have to discuss again.
320
00:15:50,480 --> 00:15:51,520
Yeah, that's right.
321
00:15:51,560 --> 00:15:51,960
That's right.
322
00:15:52,960 --> 00:15:53,400
Yes.
323
00:15:53,600 --> 00:15:54,360
All right.
324
00:15:54,360 --> 00:15:57,360
Well, we will move on from this hot topic.
325
00:15:57,360 --> 00:16:00,440
But again, thank you all for listening.
326
00:16:00,440 --> 00:16:05,040
And until next time, stay safe, stay secure and stay private.
327
00:16:05,080 --> 00:16:06,680
Thank you for listening or watching.
328
00:16:06,840 --> 00:16:10,720
If you like this episode, hit subscribe, share it with your friends and
329
00:16:10,720 --> 00:16:14,760
colleagues or jump into our community at sharedsecurity.net
330
00:16:14,760 --> 00:16:17,400
slash supporter to keep the conversation going.
331
00:16:17,480 --> 00:16:22,200
Thanks again, and we'll see you next week for another episode of Shared Security.
00:00:00,000 --> 00:00:03,200
cows will wipe out human race if it happened.
2
00:00:03,400 --> 00:00:10,440
You could have a cow wander into a electrical transmission system,
3
00:00:10,440 --> 00:00:15,680
knock out power for a large portion of the country, of whatever country it is.
4
00:00:16,040 --> 00:00:20,800
You know, as well as I do, that you go 48 hours without power.
5
00:00:20,880 --> 00:00:22,400
People are killing people, right?
6
00:00:22,400 --> 00:00:23,600
It's going to happen.
7
00:00:23,600 --> 00:00:25,760
And that's just going to cycle, right?
8
00:00:25,760 --> 00:00:28,640
Non-zero chance cows will kill humanity.
9
00:00:28,800 --> 00:00:33,760
This week on Shared Security, we pose a question, will AI kill us all?
10
00:00:34,000 --> 00:00:38,040
Well, this is one hell of a question to ask before your morning coffee.
11
00:00:38,320 --> 00:00:43,400
And suddenly everyone, including the mass media and the US government,
12
00:00:43,400 --> 00:00:46,160
is talking about this single question.
13
00:00:46,160 --> 00:00:50,400
Anthropics Dario Amade wants the frontier AI race paced.
14
00:00:50,640 --> 00:00:54,040
And a former anthropologist researcher says we will all be dead
15
00:00:54,040 --> 00:00:59,080
by the end of this decade by a real life skynet from the Terminator movies.
16
00:00:59,360 --> 00:01:01,080
Hasta la vista, baby.
17
00:01:01,080 --> 00:01:06,400
Now, Scott, Kevin and I aren't selling a robot apocalypse countdown here.
18
00:01:06,640 --> 00:01:12,600
We're asking what must be proven before AI gets more autonomy, access and control
19
00:01:12,880 --> 00:01:18,320
and whether anyone, including the AI companies themselves, will actually hit the brakes.
20
00:01:18,760 --> 00:01:20,400
Well, what do you guys think?
21
00:01:20,400 --> 00:01:23,440
Is it time to pack our things and head for the mountains?
22
00:01:23,720 --> 00:01:28,720
Or is this just more AI hype that really doesn't mean anything?
23
00:01:28,880 --> 00:01:31,280
What do you think? Hype or is it real?
24
00:01:31,280 --> 00:01:33,280
Yes and no. Yes and no.
25
00:01:33,280 --> 00:01:37,600
Like right off the bat, I believe and have long stood by the statement
26
00:01:38,320 --> 00:01:44,280
that saying something has a non-zero chance of happening is a less measurement.
27
00:01:44,840 --> 00:01:48,160
It is as useful as me saying, hey, Tom,
28
00:01:48,200 --> 00:01:52,480
you have a greater than average number of legs, right?
29
00:01:52,760 --> 00:01:57,520
Which because the average legs on humans is less than two.
30
00:01:57,840 --> 00:01:58,320
All right.
31
00:01:58,320 --> 00:02:01,320
And I know other people sitting there listening right now going, no,
32
00:02:01,320 --> 00:02:03,000
almost everybody has two legs.
33
00:02:03,000 --> 00:02:04,320
I'm like, I understand that.
34
00:02:04,320 --> 00:02:07,160
But the way averages work, that's right.
35
00:02:07,720 --> 00:02:11,120
Right. Unless anybody has the exact same number,
36
00:02:11,160 --> 00:02:14,880
the number, the average number will be lower than the biggest number.
37
00:02:15,200 --> 00:02:16,760
Right. Just math.
38
00:02:16,760 --> 00:02:19,080
So saying we have a non-zero.
39
00:02:19,080 --> 00:02:20,240
OK. Yes.
40
00:02:20,240 --> 00:02:21,880
You know what else we have a non-zero chance
41
00:02:21,880 --> 00:02:24,360
of killing the entire human race cows.
42
00:02:24,840 --> 00:02:26,800
There is a non-zero chain.
43
00:02:26,800 --> 00:02:30,040
OK. That cows will wipe out human race.
44
00:02:30,200 --> 00:02:34,600
If it happened, you could have a cow wander into a
45
00:02:34,880 --> 00:02:40,400
electrical transmission system, knock out power for a large portion
46
00:02:40,440 --> 00:02:43,640
of the country of whatever country it is.
47
00:02:43,960 --> 00:02:46,640
You know, as well as I do, we've all seen the movie Purge
48
00:02:47,040 --> 00:02:51,000
that you go 48 hours without power.
49
00:02:51,080 --> 00:02:52,640
People are killing people, right?
50
00:02:52,640 --> 00:02:56,200
I just it's going to happen and that's just going to cycle, right?
51
00:02:56,440 --> 00:02:59,360
Non-zero chance cows will kill humanity.
52
00:02:59,480 --> 00:03:04,080
So I just end with all the things happening with, you know,
53
00:03:04,080 --> 00:03:06,960
going on around now, as as Brittany just messaged me,
54
00:03:07,160 --> 00:03:08,440
because he used meat.
55
00:03:08,440 --> 00:03:11,080
I'm surprised we haven't had to be cold yet of that.
56
00:03:11,880 --> 00:03:15,080
Do we know that's not caused by some rogue AI that's.
57
00:03:15,480 --> 00:03:18,360
I'll tell you right now, it was because rogue AI
58
00:03:18,360 --> 00:03:20,640
and that rogue AI is named RFK Junior.
59
00:03:21,280 --> 00:03:24,240
Oh, sorry.
60
00:03:24,280 --> 00:03:25,840
I set you up for that one, Kevin.
61
00:03:25,840 --> 00:03:27,240
You did. I appreciate that.
62
00:03:27,240 --> 00:03:29,280
I know that here's here's my opinion.
63
00:03:29,320 --> 00:03:33,640
I think that there's two problems with this introduction.
64
00:03:33,680 --> 00:03:34,920
And I don't mean your introduction.
65
00:03:34,920 --> 00:03:36,680
I think your introduction was awesome.
66
00:03:36,680 --> 00:03:39,680
One, the idea of non-zero chance.
67
00:03:39,680 --> 00:03:41,240
Of course, there's a non-zero chance.
68
00:03:41,240 --> 00:03:44,400
We have systems that people are using to make decisions,
69
00:03:44,400 --> 00:03:48,680
even if we don't give the AI system the autonomy to do something, right?
70
00:03:48,880 --> 00:03:51,320
How many people here have seen the episode of the office
71
00:03:51,640 --> 00:03:55,040
where Michael and Dwight drive into the lake
72
00:03:55,040 --> 00:03:57,640
because the GPS told them to turn there?
73
00:03:57,920 --> 00:03:59,440
Right. Yeah. Yeah.
74
00:03:59,440 --> 00:04:04,240
You can't convince me that Hagswith or whoever
75
00:04:05,000 --> 00:04:07,640
isn't asking Claude something and having Claude say,
76
00:04:07,640 --> 00:04:09,480
you'll be wall and they would do it, right?
77
00:04:09,480 --> 00:04:10,520
Of course. Yeah.
78
00:04:10,520 --> 00:04:12,960
OK, so non-zero chance.
79
00:04:12,960 --> 00:04:14,960
Here's the other problem I have with this entire thing.
80
00:04:14,960 --> 00:04:17,560
And we keep falling for this stuff, right?
81
00:04:17,920 --> 00:04:19,000
We've talked about this before.
82
00:04:19,000 --> 00:04:21,040
I think it was Google we talked about before.
83
00:04:21,040 --> 00:04:24,920
Every single time knows Facebook, every single time
84
00:04:25,280 --> 00:04:30,200
an organization is closer to a monopoly or success
85
00:04:30,680 --> 00:04:34,040
in their field and they're being chased
86
00:04:34,040 --> 00:04:36,440
by a whole bunch of smaller competitors.
87
00:04:37,080 --> 00:04:39,600
All of a sudden, they demand legislation
88
00:04:39,800 --> 00:04:43,120
against the thing they're doing. Yeah, it's so crazy.
89
00:04:43,120 --> 00:04:45,440
Yeah. The only organization
90
00:04:45,440 --> 00:04:49,080
that is going to be helped by legislating and limiting
91
00:04:49,080 --> 00:04:52,360
our research into AI is Anthropic and Open AI.
92
00:04:53,040 --> 00:04:55,120
Period. Right.
93
00:04:55,360 --> 00:04:57,080
And I'm not saying we shouldn't have guardrails.
94
00:04:57,080 --> 00:04:58,880
I'm not saying we shouldn't have something.
95
00:04:58,880 --> 00:05:00,800
We all know my opinion of legislation.
96
00:05:00,800 --> 00:05:04,000
But every single time you hear an organization that says
97
00:05:04,040 --> 00:05:06,120
because here's what they they say part of this
98
00:05:06,120 --> 00:05:08,720
and the other part is just kind of implied.
99
00:05:08,720 --> 00:05:11,520
Hey, we've spent a lot of money building something.
100
00:05:12,480 --> 00:05:14,120
And because we've built it,
101
00:05:14,120 --> 00:05:15,480
we've proved it's possible.
102
00:05:15,480 --> 00:05:17,360
So other people are now trying to build it.
103
00:05:17,360 --> 00:05:20,960
And sadly for us, it will be easier
104
00:05:20,960 --> 00:05:23,520
for them to build it because we've proven the path.
105
00:05:23,760 --> 00:05:26,560
So what we want to do is spike the path
106
00:05:26,800 --> 00:05:28,720
so they can't come in. Yeah. Right.
107
00:05:29,080 --> 00:05:31,960
And the way they spike the path is by going to Congress
108
00:05:31,960 --> 00:05:34,360
and saying, hey, we paid you a bunch of money.
109
00:05:34,360 --> 00:05:35,360
I mean, we didn't pay you.
110
00:05:35,360 --> 00:05:38,040
We donated to your election campaign
111
00:05:38,320 --> 00:05:40,760
past this law that makes it harder for everybody else.
112
00:05:41,640 --> 00:05:42,920
Yeah. I'm tired of that.
113
00:05:42,920 --> 00:05:45,840
And I'm tired of the fact that we as a population
114
00:05:46,160 --> 00:05:48,760
keep just accepting that call.
115
00:05:48,960 --> 00:05:51,760
The fact that people I know that have.
116
00:05:51,760 --> 00:05:55,520
They could not spell AI are now saying,
117
00:05:55,520 --> 00:05:58,840
well, we should have laws protecting us from this AI system.
118
00:05:58,960 --> 00:06:00,480
I look a cool.
119
00:06:00,480 --> 00:06:02,680
And how many laws have protected us against murder?
120
00:06:03,080 --> 00:06:05,720
How many law rights? Sorry.
121
00:06:05,720 --> 00:06:08,080
I'm right. Yeah. No, no, it's a it's a good rant.
122
00:06:08,080 --> 00:06:09,320
And I agree with you, Kevin.
123
00:06:09,320 --> 00:06:11,760
I think I find it funny
124
00:06:11,840 --> 00:06:15,360
that the value of these companies
125
00:06:15,480 --> 00:06:18,240
just continue to increase every time
126
00:06:18,440 --> 00:06:21,640
a founder like Dario, whoever comes out and says,
127
00:06:22,160 --> 00:06:25,520
AI needs to be paused, needs to be regulated, needs to.
128
00:06:25,640 --> 00:06:27,480
We need to stop what we're doing.
129
00:06:27,480 --> 00:06:30,080
We have to do something about it and put guardrails in.
130
00:06:30,080 --> 00:06:32,920
And like, I'm like, well, first of all,
131
00:06:32,920 --> 00:06:35,360
the investors just keep throwing more money
132
00:06:35,360 --> 00:06:37,440
and billions of dollars at these companies.
133
00:06:37,440 --> 00:06:38,880
The more they say that.
134
00:06:38,880 --> 00:06:42,560
And then second, they're the ones creating these things,
135
00:06:42,560 --> 00:06:44,960
but they're not pausing and they're not stopping.
136
00:06:44,960 --> 00:06:46,640
They're continuing to do this.
137
00:06:46,640 --> 00:06:49,920
So to me, it's all marketing and it's about money and investment.
138
00:06:49,920 --> 00:06:50,760
Absolutely.
139
00:06:50,760 --> 00:06:52,560
The fact that they're getting more investment
140
00:06:52,560 --> 00:06:56,280
when they're literally saying government make us stop
141
00:06:56,680 --> 00:06:59,800
tells me that the investors know they're not going to
142
00:06:59,800 --> 00:07:01,120
and the law is not going to affect them.
143
00:07:01,120 --> 00:07:04,480
I have this great idea if Dario or whatever.
144
00:07:04,480 --> 00:07:08,560
And for some reason, I just see him as a Italian plumber in a game.
145
00:07:08,920 --> 00:07:12,760
Every time I hear his name, Mario, Mario and Mario Dario,
146
00:07:13,040 --> 00:07:19,240
Evil twin brother, Dario and every time I see these,
147
00:07:19,600 --> 00:07:21,720
like, I want to write back to the dude and say,
148
00:07:21,720 --> 00:07:23,520
I got this great idea.
149
00:07:23,520 --> 00:07:25,200
I got this great idea, man.
150
00:07:25,200 --> 00:07:27,960
You believe that I should be regulated.
151
00:07:28,760 --> 00:07:31,880
So give me a list of the things you want the government
152
00:07:31,880 --> 00:07:34,240
to make it illegal for AI to do.
153
00:07:34,240 --> 00:07:36,280
OK, cool. Give me that list.
154
00:07:36,280 --> 00:07:39,400
Now, tell me why you haven't stopped your own damn company.
155
00:07:39,400 --> 00:07:40,800
You know, the one you're in charge of,
156
00:07:40,800 --> 00:07:42,840
you know, the one you make the rules for,
157
00:07:42,840 --> 00:07:47,240
the one that you claim is just trying to help population
158
00:07:47,240 --> 00:07:51,560
and humanity will implement those rules, jerk, right?
159
00:07:51,800 --> 00:07:55,920
The day you show me that you have taken your proposed
160
00:07:55,920 --> 00:07:59,480
regulation and made it not policy,
161
00:07:59,480 --> 00:08:03,440
I want to see actual controls that prevent the thing
162
00:08:03,440 --> 00:08:07,120
you're claiming and everybody else is going to do.
163
00:08:07,680 --> 00:08:08,600
Show me those controls.
164
00:08:08,600 --> 00:08:10,920
And the day you show me those controls and show me
165
00:08:10,920 --> 00:08:13,760
that they're effective, then I will have an actual
166
00:08:13,760 --> 00:08:15,760
conversation with you about regulation.
167
00:08:15,760 --> 00:08:18,680
But until you show me that you're you're putting
168
00:08:18,680 --> 00:08:21,560
your money where your mouth is, because, like I said,
169
00:08:21,960 --> 00:08:24,760
you control the organization.
170
00:08:25,000 --> 00:08:27,360
That would be like me saying, you know what?
171
00:08:27,360 --> 00:08:30,040
Secure ideas has this really serious problem
172
00:08:30,360 --> 00:08:33,960
that our consultants are becoming artificially intelligent
173
00:08:34,160 --> 00:08:37,440
and they may destroy the world.
174
00:08:38,160 --> 00:08:40,160
But if the government doesn't regulate me from
175
00:08:40,160 --> 00:08:42,080
preventing them from destroying the world, look,
176
00:08:42,080 --> 00:08:44,080
my consultants are going to go crazy.
177
00:08:44,080 --> 00:08:45,400
Yeah. I'm sorry.
178
00:08:45,400 --> 00:08:47,200
We told you. My hands are tied.
179
00:08:47,200 --> 00:08:49,280
I got to be destructive.
180
00:08:49,480 --> 00:08:52,600
Only the government can save you from secure ideas.
181
00:08:53,440 --> 00:08:55,520
It's BS. I'm sorry. It's BS.
182
00:08:56,320 --> 00:08:58,440
OK, I'll shut up now.
183
00:08:58,480 --> 00:09:00,240
Scott, what do you think of all this?
184
00:09:00,240 --> 00:09:02,520
Well, I mean, I can't create as entertaining
185
00:09:02,520 --> 00:09:04,960
an argument as Kevin on whether or not
186
00:09:04,960 --> 00:09:06,720
things should be regulated.
187
00:09:07,360 --> 00:09:08,600
So I'm just going to leave it at that.
188
00:09:08,600 --> 00:09:10,800
But I think there are a lot of different paths
189
00:09:10,800 --> 00:09:14,240
by which I could severely harm the human race.
190
00:09:14,240 --> 00:09:16,240
And I'm just going to go down that path.
191
00:09:16,840 --> 00:09:17,640
Yeah, there you go.
192
00:09:17,640 --> 00:09:20,520
And say, you know, the recent,
193
00:09:21,480 --> 00:09:23,600
you know, evolutions of, you know,
194
00:09:23,600 --> 00:09:25,840
breaking out of testing sandboxes
195
00:09:26,320 --> 00:09:28,240
to me is the most credible.
196
00:09:28,240 --> 00:09:31,040
A lot of people will say, yeah, it's it's marketing.
197
00:09:31,040 --> 00:09:32,800
You know, they did it to make their
198
00:09:34,120 --> 00:09:36,400
things seem more mysterious and powerful.
199
00:09:36,760 --> 00:09:39,960
But I can think of other ways that they might have come up with,
200
00:09:39,960 --> 00:09:43,440
you know, a killer app that would grow their market,
201
00:09:43,840 --> 00:09:46,160
maybe not as fast or I don't know.
202
00:09:46,160 --> 00:09:49,040
I'm just saying I think that the most credible way
203
00:09:49,040 --> 00:09:52,520
that this stuff gets to that point of being,
204
00:09:52,520 --> 00:09:57,200
you know, a serious threat to all humanity is the way it's going.
205
00:09:58,080 --> 00:09:59,920
In the path that we've seen it going on.
206
00:09:59,920 --> 00:10:03,400
And I think, you know, some of the scariest things we could see,
207
00:10:03,400 --> 00:10:08,760
if my feeling is correct, is it would be finding paths
208
00:10:08,760 --> 00:10:11,280
to a distributed way of operating
209
00:10:11,280 --> 00:10:13,720
so you don't have a kill switch, right?
210
00:10:13,720 --> 00:10:17,680
That would it would looking back to Asimov's, you know,
211
00:10:17,680 --> 00:10:19,960
three rules of robotics, right?
212
00:10:19,960 --> 00:10:24,720
You know, did you see the XKCD cartoon about that?
213
00:10:25,080 --> 00:10:26,200
You got to find it.
214
00:10:26,440 --> 00:10:30,280
Why Asimov put his rules in the order that he did.
215
00:10:30,360 --> 00:10:32,120
It's actually the picture of the week
216
00:10:32,120 --> 00:10:36,200
and last week's security now episode from Steve Gibson.
217
00:10:36,200 --> 00:10:40,520
But basically every other combination of don't harm humans,
218
00:10:40,520 --> 00:10:43,880
obey the rules and then, you know, protect yourself, right?
219
00:10:44,160 --> 00:10:46,880
Every other combination results in bad things.
220
00:10:47,640 --> 00:10:50,720
I don't know how I'm not an expert on this stuff.
221
00:10:50,720 --> 00:10:55,480
But I think the process starts with identifying, you know,
222
00:10:55,680 --> 00:10:59,680
is this going in a direction that could, you know,
223
00:10:59,680 --> 00:11:01,360
lead to what they're predicting?
224
00:11:01,360 --> 00:11:02,960
And I think it may be.
225
00:11:02,960 --> 00:11:06,080
And I've always sort of had that gut feeling that, you know,
226
00:11:06,080 --> 00:11:08,400
it was going to start with, well, you're not going to be able
227
00:11:08,400 --> 00:11:11,800
to trust what you see on TV or on, you know, streaming.
228
00:11:12,000 --> 00:11:13,600
And that's happening.
229
00:11:13,600 --> 00:11:14,440
It's happening now.
230
00:11:14,440 --> 00:11:15,760
Yeah. Yeah. Yeah.
231
00:11:15,760 --> 00:11:17,880
So, yeah, that's that's my two cents.
232
00:11:18,040 --> 00:11:20,960
And I want to be very, Scott, agree 100 percent with you.
233
00:11:21,520 --> 00:11:25,080
I'm not saying the uselessness of the measure
234
00:11:25,480 --> 00:11:28,840
means we should ignore the fact that it's possible.
235
00:11:29,080 --> 00:11:31,080
It is absolutely possible.
236
00:11:31,080 --> 00:11:34,800
As a matter of fact, I would argue possible isn't the right word.
237
00:11:35,200 --> 00:11:37,120
I think the word should be likely, right?
238
00:11:37,360 --> 00:11:41,920
Whether it's directly due to the AI system going like,
239
00:11:42,240 --> 00:11:45,080
hey, wait, these humans suck, let's wipe them out.
240
00:11:45,520 --> 00:11:49,160
Or it's just because we've moved to something
241
00:11:49,280 --> 00:11:52,200
that either people are making decisions based
242
00:11:52,200 --> 00:11:56,640
on what AI systems are doing or we're allowing AI systems
243
00:11:56,640 --> 00:12:00,360
to control things and making judgment calls
244
00:12:00,360 --> 00:12:04,880
based on incorrect information, but we're making them at speed.
245
00:12:05,080 --> 00:12:06,720
Right. Because that's that's the thing.
246
00:12:06,720 --> 00:12:10,960
Look, I'm dull, but I can only be dumb so fast.
247
00:12:11,400 --> 00:12:15,720
Right. AI could be dumb just like me way faster.
248
00:12:15,880 --> 00:12:20,240
And if I take my level of dumb and combine it with the speed
249
00:12:20,280 --> 00:12:23,480
of AI giving me more dumb, we have an issue.
250
00:12:23,760 --> 00:12:27,920
So yeah, I totally think it is possible
251
00:12:28,000 --> 00:12:32,600
and likely that we, even if we don't wipe out humanity, right?
252
00:12:34,320 --> 00:12:38,040
The idea that we cause significant harm
253
00:12:38,320 --> 00:12:42,000
directly due to the usage of AI, not
254
00:12:42,360 --> 00:12:45,080
data centers sucking up all the waters and stuff like that.
255
00:12:45,400 --> 00:12:48,480
But directly from the usage of AI, I took AI
256
00:12:48,640 --> 00:12:52,520
and hosted up to this nuclear power plant and it went right in the meltdown.
257
00:12:52,520 --> 00:12:55,280
Or it's even less obvious than that.
258
00:12:55,280 --> 00:12:59,160
I think, you know, that method I said of getting to a distributed operation.
259
00:12:59,680 --> 00:13:03,000
What about what about the botnets that have taken over all the old routers?
260
00:13:03,680 --> 00:13:04,880
Right. Hmm.
261
00:13:04,880 --> 00:13:08,400
Isn't that a logical way that they might go, right?
262
00:13:08,400 --> 00:13:11,360
Rather than first attacking a nuclear reactor?
263
00:13:11,720 --> 00:13:13,480
Well, I'm not saying attacking a nuclear reactor.
264
00:13:13,480 --> 00:13:17,240
I'm saying we give it control to the nuclear reactor and it makes a mistake.
265
00:13:17,280 --> 00:13:18,720
Yeah. Oh, I see. Yeah. Yeah.
266
00:13:18,720 --> 00:13:22,400
Yeah. Accidentally, I guess I'm more of a realist
267
00:13:22,400 --> 00:13:25,720
that like we don't need AI to destroy humanity.
268
00:13:25,720 --> 00:13:27,320
Like we're already doing pretty well.
269
00:13:27,320 --> 00:13:32,080
It's like we're like, yeah, like literally like we got climate change.
270
00:13:32,080 --> 00:13:35,000
We have, you know, I mean, pick your poison, right?
271
00:13:35,000 --> 00:13:38,880
Like I'm actually more fearful of, you know, World War Three
272
00:13:38,880 --> 00:13:40,680
and all of us launching nukes at each other.
273
00:13:40,680 --> 00:13:45,480
And there definitely is the end of humanity and AI won't even be involved in it.
274
00:13:45,480 --> 00:13:48,040
Right. And I see this all over the media and the news.
275
00:13:48,040 --> 00:13:49,720
First of all, the odds of this happening.
276
00:13:49,720 --> 00:13:51,880
Yes, there's always a chance and all these things.
277
00:13:51,880 --> 00:13:53,760
Well, it's going to happen in five, 10 years.
278
00:13:53,760 --> 00:13:55,160
Like who really knows?
279
00:13:55,160 --> 00:13:57,000
Like most likely not.
280
00:13:57,000 --> 00:13:59,600
I mean, and I think it is kind of overhyped,
281
00:13:59,600 --> 00:14:03,960
but I think we need to direct our attention towards the things that we can control.
282
00:14:05,320 --> 00:14:11,120
Right. Like and realize that this AI hype is really all of a lot of it
283
00:14:11,120 --> 00:14:14,680
is built around money and investments.
284
00:14:14,680 --> 00:14:20,160
And we probably don't have time to talk about the whole US versus China thing.
285
00:14:20,160 --> 00:14:23,600
And Trump saying is like, well, we can't have guardrails.
286
00:14:23,600 --> 00:14:24,760
We got to beat China.
287
00:14:24,760 --> 00:14:26,160
Guys, China's already won.
288
00:14:26,160 --> 00:14:27,640
Really sharp president.
289
00:14:27,640 --> 00:14:29,680
No, seriously, China's already ahead of us.
290
00:14:29,760 --> 00:14:31,400
Like we should just admit that.
291
00:14:31,400 --> 00:14:33,960
But we need to focus back on what we can control.
292
00:14:34,240 --> 00:14:37,640
And it may not be AI turning into the next Skynet.
293
00:14:37,840 --> 00:14:38,680
Just my opinion.
294
00:14:38,680 --> 00:14:39,800
Yeah, priorities.
295
00:14:39,840 --> 00:14:43,280
This one is a hotly a hot topic.
296
00:14:43,280 --> 00:14:47,440
And we all have very, very strong opinions about it.
297
00:14:47,440 --> 00:14:49,240
But love to hear from our audience.
298
00:14:49,240 --> 00:14:50,040
What do you think?
299
00:14:50,040 --> 00:14:53,680
Do you think this is the end of the world as we know it?
300
00:14:53,680 --> 00:14:56,240
And all of humanity will be wiped out in five years?
301
00:14:56,240 --> 00:15:00,880
Or do you think it's a more hype and more marketing and more about money?
302
00:15:00,920 --> 00:15:03,480
And you put the REM song in the background.
303
00:15:03,480 --> 00:15:06,120
I could maybe on the audio version.
304
00:15:06,200 --> 00:15:10,400
But YouTube will ban me for life if I use the copyright.
305
00:15:10,400 --> 00:15:12,880
So hopefully me seeing it just now won't get you banned.
306
00:15:12,880 --> 00:15:15,080
But, you know, yeah, well, I may have to edit that out.
307
00:15:15,080 --> 00:15:15,840
Scott, thank you.
308
00:15:18,000 --> 00:15:19,160
Do you have any questions to anyone?
309
00:15:19,160 --> 00:15:20,360
Yes, yes, that's right.
310
00:15:20,360 --> 00:15:21,640
Please let us know in the comments.
311
00:15:21,640 --> 00:15:24,760
We've been getting some great comments on the on the YouTube channel recently.
312
00:15:25,080 --> 00:15:29,000
Even on past posts, I saw a really good one about remember the topic
313
00:15:29,000 --> 00:15:31,720
we had on AI boyfriends and girlfriends.
314
00:15:31,720 --> 00:15:36,080
And someone left a very good, well thought out comments about that.
315
00:15:36,120 --> 00:15:40,040
And I was good about, like, we should be really focused on human
316
00:15:40,040 --> 00:15:43,800
relationships with each other and not so much relationships
317
00:15:43,800 --> 00:15:47,480
with AI systems that, well, will never be human.
318
00:15:47,680 --> 00:15:48,800
There's some good studies on that.
319
00:15:48,800 --> 00:15:50,480
And maybe we may have to discuss again.
320
00:15:50,480 --> 00:15:51,520
Yeah, that's right.
321
00:15:51,560 --> 00:15:51,960
That's right.
322
00:15:52,960 --> 00:15:53,400
Yes.
323
00:15:53,600 --> 00:15:54,360
All right.
324
00:15:54,360 --> 00:15:57,360
Well, we will move on from this hot topic.
325
00:15:57,360 --> 00:16:00,440
But again, thank you all for listening.
326
00:16:00,440 --> 00:16:05,040
And until next time, stay safe, stay secure and stay private.
327
00:16:05,080 --> 00:16:06,680
Thank you for listening or watching.
328
00:16:06,840 --> 00:16:10,720
If you like this episode, hit subscribe, share it with your friends and
329
00:16:10,720 --> 00:16:14,760
colleagues or jump into our community at sharedsecurity.net
330
00:16:14,760 --> 00:16:17,400
slash supporter to keep the conversation going.
331
00:16:17,480 --> 00:16:22,200
Thanks again, and we'll see you next week for another episode of Shared Security.
