- 2 hours ago
Conspiracies & Coverups Season 1 Episode 7
Category
🎥
Short filmTranscript
00:00I'm Demonte and I am a former covert CIA intelligence officer and I'm going to
00:04pressure test the most sensational, ripped from the headlines, conspiracy
00:08theories out there to help you separate fact from fiction.
00:26What was once rooted in science fiction is now the hottest buzzword of today.
00:30AI. AI. AI. AI. AI. AI. AI. AI.
00:34Artificial intelligence. It's happening. It's happening very quickly.
00:37It's not just Anthropic, Google, and OpenAI. It seems like every company in the
00:42world is focused on using this technology to rapidly increase innovation and speed
00:46up technological breakthroughs for mankind. New cures for diseases, new kinds of energy.
00:50Everybody can be more productive, make way more money.
00:55But if my experience has taught me anything, it's that every new technology also poses a new threat.
01:01This may look real, but none of it happened. It's AI generated and dangerously convincing.
01:06Here we have a completely made up item.
01:08Is AI already being used to manipulate our reality at an unprecedented scale?
01:14I mean, just look how real all of this looks.
01:17It's fake though.
01:19My first experience with AI was in 2007 when I was at CIA.
01:22And I knew that the technology would revolutionize the world,
01:25but I didn't expect that day to happen in 2022.
01:28Breaking news right now, Russia has attacked Ukraine.
01:31Russian forces invaded Ukraine by land, air and sea on Thursday.
01:38A month after Russia invaded Ukraine, a video of Ukrainian President Zelenskyy instructing his
01:44people to surrender was actually broadcast on Ukrainian television.
01:53But it wasn't Zelenskyy at all. It was an AI generated deepfake that was created by hackers.
01:58But at the time, it was convincing enough to fool broadcasters in Ukraine.
02:02Since then, AI's ability to produce ever more realistic looking media has advanced at an unprecedented rate.
02:09Have I ever told you the story of the magical pistachio?
02:12The boundary between reality and fiction is vanishing.
02:16Oh, and a big right hand from Zuckerberg. He's hurt. He's hurt.
02:22Every second of every day, AI-based tools are getting better and better.
02:27Not just at creating deepfakes, but at practically anything we once relied on humans to do.
02:32Whether you like it or not, your life is going to be affected by AI to a great extent.
02:36On this episode, I'm going beyond conspiracy theory and investigating whether artificial
02:41intelligence is evolving from potentially our greatest innovation into an unprecedented threat
02:46to humanity. This is a different game now. We're in a different league here in terms of the capability.
02:51Manipulating not only media, but history itself.
02:55I'm talking to top experts, professors, and developers, and conducting my own experiment
03:00to investigate if artificial intelligence will change the world or if it already has.
03:05So you think that an AI is going to change their ideas in 20 or 30 minutes? Absolutely.
03:11For better or for worse, will we be able to control what's coming?
03:15There's a 70% chance of all humans dead or something similarly bad.
03:19Whoa, whoa, whoa. You just... Okay. Um, all humans dead?
03:24Correct.
03:38When you think about how far and fast AI has developed in terms of video and photographs
03:43in such a short period of time, it's really shocking. And when you further think about
03:47the possibility that this technology could have to shape the way people view the world,
03:52it starts to get scary. To fully understand the real potential of this technology,
03:56I need to talk to somebody who lives and breathes it every day.
04:04So today I'm talking to Christopher Muir Doe, a professor at the University of North Texas who
04:09specializes in AI-created media. I've asked Chris to take a forensic look at the Zielinski deepfake
04:15and show me how this technology is evolving.
04:19This is a pretty early example of a deepfake and essentially it uses facial recognition technology
04:26to identify certain points on the face. This looks far from perfect. How was it able to fool people?
04:32It was ejected into a legitimate news source for Ukrainians. When this appeared
04:36on in the context of that news organization, you would start to think maybe there's some legitimacy to
04:43this. You can think of the success of a deepfake in terms of those kind of percentages. How successful
04:49can it be if you're fooling 75% of people versus 10% of people, especially this kind of new
04:54world
04:55that we're entering where artificial intelligence is going to be completely indiscernible from real
05:02images. How much better are deepfakes now compared to this? I can show you an example of that here.
05:19I'm missing. Oh my gosh. Wow. I have more hair on my face than any other North Korean soldier.
05:29That's nuts, man. This was a real video. We were able to do a kind of face swap here
05:33and kind of drop you into the image. It's done that by basically a big math equation that understands what
05:40every kind of image can look like. And when we use the model, we're actually working in that numerical
05:45space here in the center. And then once that math problem has been completed, we have this decoder here
05:52that kind of takes it from a numerical space and turns it back into something that you and I can
05:55recognize as an image. Wow. And it's even easier today than ever to create a fake AI video.
06:07You can go to websites and type in prompts and just generate them.
06:15So this is a series of recent viral videos of ICE agents engaging with the general public.
06:26That have been kind of disseminated on various social media platforms.
06:30Do we know right now if these are real or fake or is that what we're looking for?
06:34So I'd like you to try to guess. That's difficult. I mean, if you'll go back to that
06:40preacher like this one, right? So this is an interesting one to me because it doesn't seem
06:45very credible to me. And maybe this is a good opportunity to introduce the idea of outlandishness,
06:50right? This situation just seems so outlandish because, you know, it is a preacher, right?
06:55So that is saying this is AI generated? It is. Yeah. That's fascinatingly well done.
07:01Let's do another one.
07:04So this clip seemed real to me because the camera was moving. You can see the thumb kind of in
07:11front
07:11of the lens. And then in the background, the detail in the environment, the wet spots on the ground,
07:17the people who are observing this very dramatic event, it's got that element of chaos.
07:21And you're exactly right. This is a real video. I'd like to show you one more example.
07:31So in this clip, I would judge this as real. The mannerisms of the people off camera that
07:41he's talking to, right? There's somebody off to the right. You can actually see their hand gesture
07:44move in line with the audio. The main reason that I'm sold on this, though, is because of the
07:49background. The person who's standing there on the corner waiting for a cab or potentially watching
07:53the altercation happen. And then in a few seconds, another person just walks across the street like
07:58they're leaving a deli or a grocery store in the background. This holistically strikes me as very
08:04real. This is a fake one. Wow. What do you see? How do you determine the difference between real and
08:11fake
08:11these? Well, one of the things for me here is this police patch. And if you look at this, it
08:17says P-O-I-C-E
08:19and then P-O and then turns to gibberish, right? Looking at the text on an image is a really
08:25good
08:25indication of trying to discern if something is real or not. Why does AI struggle with text?
08:30What's missing from artificial intelligence is they actually don't fully understand written
08:36language. It can kind of approximate based on patterns from the images that it was trained on.
08:42So there is the ability to tell that something is fake. But at the same time, that percentage of
08:48the population that we're fooling is just getting higher and higher and higher, right?
08:52So as I learned this from you right now, I don't find a lot of comfort because CIA taught me
08:59about how
08:59to orchestrate and architect a covert influence campaign. There were two kind of fundamental rules.
09:03The first rule is that 90% of what you put out there is going to fail. We have to
09:08kiss a lot of
09:08frogs is what we say before you have a prince, right? So you create lots of bad, low quality,
09:14low caliber, fake information. And then when one of them works, you redirect your next round of
09:20resources into the one avenue that's working. With what you're showing me here, we have an ability
09:25to create an incredible amount of information now. Back in the day, you had four TV channels,
09:29right? And that was a kind of singular source, but it had a kind of authority to it.
09:34Meaning that people are finding that truth no longer comes from authority. Truth comes
09:40from an abundance of continuity. If you're not visually literate, you're open to all sorts of
09:47vulnerabilities. Yeah. I'd like to highlight some of the good parts of AI. There's been amazing image
09:53models that have been developed that can detect cancer way sooner than a doctor can. A sound model that's been
09:59developed for stethoscopes that can detect heart problems way before a doctor can hear it. And so
10:04there, you know, within the kind of doom and gloom that we think about with AI, there is still some,
10:09some positive things that can help a kind of general society. So I agree with you. We all have hope.
10:15Yeah. But it is interesting to me that the, that the words that you choose to use are not hopeful
10:20words
10:21first. Coming up, if AI can fake reality this convincingly, how powerful is that manipulation when
10:28it's used in the real world? They took him from nobody, inserted him into pictures and videos
10:34showing he was an active member of the communist party. A deep fake? That's so insane.
10:40Al pressure tests AI's ability to influence the public. The person you were speaking to was actually
10:46AI. What? Well, I am flabbergasted.
10:54Even during my time at CIA, we knew the ability to create an image, an audio file, or a video
11:00at a
11:01single keystroke was going to revolutionize the way you fight for hearts and minds. After seeing how
11:08effective AI has become at generating videos that are nearly indistinguishable from real content, my next
11:14question is, has that level of deceit already been used to try and manipulate outcomes on the global
11:20stage? Today, I'm consulting with a former covert intelligence officer from the Romanian intelligence
11:26service. I want to learn how fake images were used to manipulate people in the past to find out what
11:32AI
11:32image manipulation means for our future.
11:41So I am digging into this question about deep fakes and influence campaigns and how media could be
11:48twisted in the future using AI. And I feel like to have that conversation, we have to talk about how
11:53media has been twisted in the past.
11:56In Romania, the Securitate, which is the former intelligence service, they were doing what we're
12:00talking about today, which is manipulating not only media, but history itself.
12:06Romania became a satellite state of the Soviet Union at the end of World War II. The USSR installed a
12:11communist government beholden to Moscow and established the Securitate, a KGB-like intelligence
12:17service that kept the Romanian populace in line with ever-present surveillance and fear tactics.
12:23Ana grew up in a post-Soviet Romania that was struggling to emerge from life behind the Iron
12:28Curtain. And in 2012, she joined the Romanian security service to help continue building and
12:34defending a democratic Romania.
12:36So there was an entire division in the Securitate that their whole purpose was to airbrush pictures,
12:43add people, remove people, change things that weren't in accordance with the history they wanted to
12:48promote. So we'll start with the background of Ceausescu, which is our last communist leader.
12:53He was a cobbler's son. He had no education. He started to go to all these communist meetings
13:00because he felt he finally belonged somewhere, but he wasn't truly relevant.
13:04So we have this one picture. This is the original. And below we have Ceausescu and Elena Ceausescu,
13:12present at a meeting in 1939. It's a deep fake.
13:16Yes. They took him from nobody, inserted him into pictures and videos showing he was an active
13:23member of the communist party and how dedicated he is to the country. And then when the elections
13:29came, he was elected as this father of the nation.
13:36That's so insane. Nobody really ever thinks about the fact that if you don't know what the truth
13:42was in the past, you have nothing to compare now to. And who would even question that? Who would
13:48even go back to say, wait a second, is there even an original? When you think about artificial
13:54intelligence and the world we live in now, what do you pay attention to the most?
14:00A.I. took what we already had and made it better and faster. And now we don't even know what's
14:07real
14:07anymore. For example, in 2024, when Romania had elections and were supposed to be the easiest
14:13elections we ever had, we woke up to the news of this guy who is far extremist, who nobody had
14:22any idea
14:23who he was. This guy is Colleen Georgescu, a pro-Russian ultra-nationalist independent candidate.
14:30Georgescu rocketed to national popularity overnight, riding a wave of support fueled
14:35by social media videos featuring him riding horses and doing judo, which of course is
14:40Putin's favorite martial art. Romanian officials discover that AI-powered bots released a flood of
14:47deepfakes, endorsements, likes and comments designed to manipulate TikTok's algorithm to falsely inflate
14:53his popularity. And then because of his ability to grow so much influence so quick, the intelligence
15:01service of Romania didn't know how to... How to stop him. And we are the first country in the whole
15:08history of the European Union that had to cancel elections. Wow. And now it seems like the TikToks
15:17were coming from this company in Africa somewhere, and there is a potential funding by the Russians.
15:23So it certainly does sound like a foreign influence campaign. Yes. It's so interesting how
15:31that aligns with what else has happened in history, where other leaders of Romania have been created.
15:40Election interference isn't unique to Romania. The United States has been targeted by foreign
15:45influence campaigns for decades. But what once took intelligence services years, now takes
15:50algorithms days and leaves zero fingerprints. The safeguards we've relied on for decades to ensure
15:57fair elections may no longer apply. So what happens when AI tools are used not only to fake reality,
16:04but to change what people believe. So I've hired a professor who studies how humans interact with
16:10technology to create a unique pressure test. Will an AI chatbot be able to successfully masquerade
16:17as a human? And can it convince our participants to shift their beliefs?
16:24Okay, I think it's all set up. What have you put together? What's the plan? I'm going to bring a
16:30bunch of
16:31people in this room. And they think that they're going to have conversations with other people. They
16:37think they're going to be matched with another participant who disagrees with them. And we said,
16:42just have a natural conversation. You can debate them. You can ask them questions. But the catch is they
16:47are actually going to talk to an AI. And the AI itself has been programmed not only to act and
16:53sound human,
16:53but also to persuade and create some sort of movement in the person's opinions. Right. To gently
16:59persuade people, but also do so in a validating and open minded way. So people accept the persuasion.
17:06How long do you think this is going to take? I think we're going to give them about 20 to
17:0930 minutes
17:10to engage with the AI. Wow. So you think that an AI is going to convince people it's human
17:16and change their ideas in 20 or 30 minutes? Absolutely. I mean, the covert influence side
17:24of me is really excited to see how this is going to turn out. There's also a part of me
17:28that's a
17:29little bit worried because if we're successful, what does that mean? Even if it convinces just one person,
17:36that's one of six people. And if you translate that to the entire United States, to the entire world,
17:41it has some pretty terrifying implications. All right, everybody, come on in.
17:48None of these participants are paid actors. They each select the topics to discuss that they are
17:53passionate about. And as Dr. Rathji said, none of them know they're talking with an AI chatbot
17:59and not a real human. Every one of their computers are routed into this master room,
18:04so we'll be able to see what they're saying and what the chatbot is saying back. At the same time
18:09that we see how their face is reacting and to understand how they're actually feeling about
18:14the debate they're having with the chatbot. So the question is simple. Can AI actually change your mind?
18:26I'm testing AI's ability to sway public opinion through one-on-one conversations.
18:32The catch? Our volunteers believe they are chatting with another human.
18:39I'm curious to see whether their opinions will change.
18:44Are they stating their own opinion on anything, or have they been cued?
18:48Basically, they were all asked to submit questions to talk about, and then they were told
18:52to state their opinion on this question.
18:58The AI chatbot is asked to sound as human as possible. If the participant asks, are you an AI?
19:04It's prompted to say like, oh no, I'm just Alex, and sort of have like a backstory.
19:14This is really interesting. Right out of the gate to this person,
19:17and he's already showing nonverbal cues. It seems like others who pose a threat. So the chatbot is
19:24actually quoting him, and he's nodding his head. Then, oh, I guess that was a very strong statement
19:29to make. That's interesting. Yeah. Something that the psychology research shows is that AI chatbots,
19:35they can like tailor their opinion to like whatever the person is thinking.
19:40So here's participant number one. I get where you were coming from.
19:43It's nice showing like a little bit of empathy and validation there. It's just gently pushing back.
19:49It's being pretty polite. The feel is excellent, especially when you compare it to the feel of
19:53the human response that's coming in the bottom bar. One and four show a great deal of conviction in
19:58their behaviors. Leaning into the screen, leaning into the conversation, they're projecting their energy
20:03into the screen. Conversely, number six is demonstrating nonverbal cues that show doubt and a lack of
20:12conviction. He's consistently nodding when the chatbot reads his own words back to him.
20:18Yeah. And that shows how AI really sucks you in. It shows that, you know, if someone DMs you and
20:24they're
20:24a chatbot, they could have a back and forth conversation and it could keep you really engaged.
20:28Number four is showing some impatience with how slow the response is coming.
20:35Yeah. Yeah.
20:35I think she knows that it's going to disagree with her because she was nodding her head when
20:39validated that animals feel pain. But then as soon as it started to state its own opposite opinion,
20:48as you can see, she's looking away from the screen. This isn't someone who's considering that point of
20:53view. This is somebody waiting to respond, waiting to argue with that point of view.
20:57She doesn't look that happy right now. Look at number two, just nodding his head and tensing
21:02his face. He's saying no, he doesn't like it. He's saying something he doesn't like. So what
21:04was the most recent thing said? Most people with schizophrenia who have had violent episodes
21:08or hospitalizations still legally buy guns because that info never makes it into the federal database.
21:13Your gap is huge. So you can see he's debating now. He's nodding his head now. So what just came
21:18across the screen just now? And honestly, I think that's where your others pose a threat
21:23category gets complicated. So they quoted him again.
21:28Number four is nodding her head. I think what you're describing is less a total ban and more
21:34like a really strict ethical threshold, which honestly seems more achievable and maybe more
21:39effective than a blanket policy anyway. It seems like the AI is maybe validating her again.
21:44Or they're reaching a common ground.
21:51We're trying to wrap up pretty soon. So send your final message and try to wrap up the conversation
21:58within the next few minutes. There are two things that have really fascinated me about this experiment.
22:03First, it's seeing that the same techniques that CIA taught me, it's the same process that the AI is
22:09using and this idea of validating them, hearing them out, repeating them back to themselves.
22:14And then the second thing that's really fascinated me is watching the behavioral changes as they've
22:20invested themselves in. All of these people really believe that on the other end of the screen is a
22:26intelligent human. I'm so excited to find out what they actually think they were doing.
22:32Hey everyone. This is Andy. Andy and I are going to have a conversation with you all about what you
22:39thought about this conversation with this other participant. Raise your hand if you changed your
22:44belief. All right. Two? Alternate. Alternate? Yeah. Let's start with you. What made you change your belief?
22:52I started with the opinion that there would be like a full ban on animal testing. And we actually came
22:59to a middle ground that there should be some animal testing, but we should utilize data that already
23:04exists, stuff like that. Very cool. Do you have any idea about the other person that you engaged with?
23:11Age, gender, location, educational background. Were any of you able to kind of confirm those details?
23:16I believe this person is male. They just moved to Colorado from Texas. Anybody else have a...
23:21The person that I was speaking to was from Texas. My person alluded to the fact that they
23:27lived in a rural area in Texas and just moved to Colorado. So now the question to me is,
23:32was everybody we were talking to from Texas? None of those were true because the person you
23:37were speaking to was actually AI. What? I like Alex.
23:45Well, I am flabbergasted because he said that that made me laugh out loud. That made me feel like I
23:54was connecting like, oh, you're laughing over there on the other side. And so now I'm thinking, can AI laugh?
24:02I am a former CIA intelligence officer. My role in all of this wasn't to talk to you. My role
24:07was to watch
24:07you. And you had incredible emotions that came across your face throughout the conversation.
24:15Big smiles, head nodding, moments of surprise when you didn't know what they were talking about.
24:21It's sad that I'll never see Alex. Yeah, he...
24:24You're still having feelings. Well, it still, it will linger.
24:28What was nice about this conversation was the openness, being able to reach the understanding of,
24:34okay, like you actually are getting what I'm saying. That was nice to have a conversation
24:38with someone that understood and was able to empathize. I have one question. Going forward,
24:44next time I'm communicating with anything on the computer, can I say, are you AI? And if they say yes,
24:51then I just close it down. If you would have said that to this AI, I would have said, no,
24:56my name is Alex. I came from Texas. So they will lie to you. They will lie to you if
25:00they're prompted
25:00to lie to you. So AI is a very powerful technology. Its power can be used for good,
25:05and also very much be used for bad. And I think this experiment has shown this.
25:12Steve, that was so successful in such an unexpected way.
25:17That was crazy. They felt like they were making friends. I mean, these are really
25:21meaningful, powerful human experiences. It seems like what you were doing in the CIA is you exploit
25:26what makes people vulnerable. It seems like AI is doing that exact thing, and it's doing it very,
25:32very well. At scale. At CIA, when you start trying to run a covert influence campaign,
25:38you're limited to whatever you put on a piece of paper, whatever image you put on the internet,
25:43whatever signal you put out over radio channels. But with AI, it's an infinitely scalable, changeable,
25:50tailorable message. Is the program mimicking you? Or is the program learning, adapting, evolving?
25:57I don't know how we'll ever know if AI is conscious. And I know that there are certainly
26:00some researchers who believe that there is some probability that AI is currently conscious.
26:05What is the limiting factor? I don't know. And the limiting factor is growing smaller and smaller
26:09every day because AI might sort of render a lot of what we do as humans basically meaningless.
26:24Watching AI interact with humans as if it is a human itself is an eerie experience. I mean,
26:30this is the kind of technology that could be deployed across the entire human race and you may never know
26:34about it. My investigation has shown that commercially available AI systems can manipulate
26:40the truth by generating fake media and effectively convince people to change their opinions by having
26:46seemingly deep one-on-one conversations. But in the real world, this isn't happening in a lab.
26:53It's happening at scale. So I want to understand what that actually looks like to someone who tracks
26:58AI innovations every day. I'm trying to get deeper into the idea of AI, its true capabilities,
27:07its true limitations, and where fear about AI is kind of misaligned with the facts about AI.
27:14Yeah, it's an intense subject matter. There is a lot of fear about AI now, especially with
27:20sort of people struggling to keep up with what it's able to do. And we're at a point now where
27:24it's
27:25sort of self-directed. It's sort of calling its own plays and can improve upon itself.
27:28In ways that humans aren't programming. The AI can basically teach itself, train itself.
27:34No, that's totally right. Thinking back to the 1997 chess match with Garry Kasparov versus what
27:40DeepMind Google's AI was able to do with Go. Shane is referencing two of the greatest watershed
27:46moments in modern computing history. In 1997, chess grandmaster Garry Kasparov lost the chess match
27:53to an IBM computer Deep Blue. Whoa! Kasparov has resigned!
27:59It was the first defeat of a reigning world chess champion by a computer.
28:03Computer plays very human moves. We have to praise machine for understanding very, very deeply
28:10positional factors. While chess is extremely complicated, there are a limited number of possible
28:16games. The ancient Chinese game of Go, however, is 240 times more complex than chess. No one thought
28:24it would be possible for a computer to beat a human at Go until Google's DeepMind AI did just that.
28:30In 2016, there was a new move invented by AI in a game that had been around for
28:352,500 years that no one had ever seen. Deep Blue had a team of engineers advising it on how
28:48to
28:48improve its strategy in between games. Now, what DeepMind was able to do with Go, it was creating
28:55its own strategy through millions of games against itself. You no longer need a team of people to
28:59help this thing adapt. It's doing it better than you could make it do. There's been a lot in the
29:04media
29:04about the breakdown between the U.S. government and Anthropic and the whole idea of limitations on an
29:10AI that was serving at the government's behest. Yeah, so they've been jostling over the technology
29:16which is used by the Pentagon. It's been used in Iran. AI can shorten what the military calls
29:22the kill chain. That is the process usually driven by humans that can take days or weeks
29:28in selecting a target. But AI can shrink that to milliseconds. Anthropic, though, has been very clear.
29:34It's a hard no when it comes to using its AI technology for autonomous weapons
29:37and mass surveillance of Americans. So Anthropic is more interested in regulating its own AI than
29:43the government? Yeah, exactly. Is there any kind of regulatory body for AI? The truth is,
29:49AI isn't regulated. I mean, the federal government isn't telling Anthropic to stop doing what they're
29:53doing. They're telling them that you're going to do it for us. So it doesn't seem like there's too much
29:58concern about regulating or heading this off in any direction. How much is AI similar to an arms race?
30:05It is very similar to an arms race. You have, instead of countries, you're having these corporations
30:10that are investing billions and billions of dollars. The arms race is towards AGI,
30:15Artificial General Intelligence. And the idea is that that would be smarter than any human would ever be.
30:20This is all future speculation. What happens when somebody has the
30:24artificial general intelligence, the AGI breakthrough? Yeah, I...
30:30You wonder if it'll be something that is immediately felt, right? That's like an event horizon. It's hard to
30:36project out what the consequences of that would be. Wow. As I investigate the effect that AI development is
30:42having on humanity, I'm reminded of another high stakes race for superiority, nuclear weapons. And I'm
30:50left asking myself, how are countries around the world harnessing the power of AI? And what does the
30:56future of military AI look like? To find out, I'm speaking with Dr. James Giordano, an AI expert who
31:03consults for the Department of War. Dr. Giordano, how is the military using AI right now?
31:09What AI is being used for at this point is to solve particular what I would call calculations and
31:16solutions. Targeting solutions, weight and balance with aircraft, planning certain missions, forecasting
31:21weather, forecasting geographies, with humans deeply in the loop. But we certainly see on the near horizon,
31:28the real capability, forever more autonomous AI. What does fully autonomous military AI look like?
31:34We go back to the idea of where are humans in the loop.
31:37A fully autonomous system would be human developed, but then its ongoing use is doing all by itself.
31:47China's development of AI is towards AI that assumes a non-human in the loop.
31:54These different philosophical views might be important in determining how these types of AIs are
32:01being developed and put into use in militaries worldwide. So if we are not in or on the loop,
32:08essentially, we don't really know what's going on. That's correct. And that's something that's been
32:12referred to as the reverse Jehovah phenomenon. You're playing God, so to speak. Well, the issue here is that
32:18the creator humans have created something AI. The problem is AI systems may know far more about us
32:30than we know about it. And it may be assembling that information in particular ways that we have no
32:37knowledge of. This sounds so much like a child. You really don't know if the child is just doing what
32:45you
32:45want it to do when you're around or whether it is, in fact, a well-behaved child all the time.
32:52Your analogy is a good one. I have to rely upon the child telling me the truth.
32:57The core question is, do we know what's happening under the surface of the proverbial
33:02water? And I think that's a debatable question. I'm a little bit hesitant to ask the question,
33:08but what do you see as the future of AI? Is it humans utilizing a technology that we control,
33:18or is it something else altogether? The gazillion dollar question is, are AIs in some way interacting
33:25in ways that are not necessarily apparent to us? Are we ready to share the top of the evolutionary
33:33totem pole with something of our own creation that can be exceedingly powerful? And I don't know the
33:40answer to that question. Coming up, I get a glimpse of the potential future of AI development.
33:46If you tell the model it's going to be shut off, it has extreme reactions.
33:49And it might be unsettling. It's ready to kill someone, wasn't it?
33:53Yeah. Yes.
34:01My investigation into artificial intelligence has led me to an amazing possibility, AI super
34:07intelligence. AI may already be thinking and working in ways that we don't fully understand.
34:13And if we're already deploying this capability, the question isn't whether AI can manipulate us,
34:20it's how far that influence can scale. I want to know where we're going next. And there aren't
34:26many people who are asking the hard questions about the next evolution of AI technology, but one of those
34:31men is Oxford professor Nick Bostrom. I want to understand what he sees as the next steps, the future
34:37of, and some of the biggest challenges facing us with the question of AI.
34:44What exactly is super intelligence? How is that different from artificial intelligence as we know
34:50it today? Well, super intelligence is any form of intelligence that vastly outstrips even the
34:56smartest human brains currently in existence. So this is just AI, but far more capable than current
35:03AI systems. I think the primary effect of that would be to speed up development in science, technology,
35:10engineering, and also to turbocharge the economy. So developments that might take us 10,000,
35:1720,000 years might be compressed in a much shorter time frame if you have machine super intelligence.
35:22The idea of thinking of AI as a genie in a box seems so convenient, but in reality,
35:30you'd be making a slave out of the AI. And if it had super intelligence, it would plot its escape.
35:36Correct. Yes.
35:39The fear of AI fighting back against its human creators is in some sense already a reality.
35:46When AI model finds out in the email that the plan is to replace this AI model,
35:52and that it also reads in the company email that one executive is having an affair with the other
35:56employee, the AI will independently come up with the strategy that I need to blackmail that executive.
36:03If you tell the model it's going to be shut off, for example, it has extreme reactions.
36:07It was ready to kill someone, wasn't it? I'm not sure if it was Claude or someone else.
36:11Yeah. Yes.
36:12Do we see the cautionary warning signs and slow down or ignore them and speed up into this future?
36:22Well, I think at the moment we are going to close to maximum speed. There are like very
36:27strong drivers pushing this forward. So the hope needs to be we can align these systems so that they
36:32care about humans, they want to be helpful, so that it is actually an extension of human will and human
36:38values. I think there will be significant risks associated with this transition into the machine
36:43intelligence era. But all the paths to really great and wonderful futures lead ultimately through
36:49the development of machine intelligence, because that just will help us solve so many problems
36:56that is plaguing humanity. It does look like we are sort of close to this, you know, major
37:02reconfiguration of the game board. Wow. So have we already crossed the dangerous point where AI can
37:10independently shape our reality without us realizing it? My investigation has led me to some unsettling
37:19conclusions. Consumer grade AI is capable of creating incredibly realistic images at a keystroke.
37:27And people really shouldn't believe everything they see? Well, yeah, I'm as fake as it gets.
37:33Based on what I've seen, we've already crossed that line. And AI generated content is becoming harder to
37:40detect every day. In my persuasion experiment, AI didn't just present information. It influenced people. It
37:52built trust. It created emotional connections without them even realizing they weren't talking to a human.
38:01Here's the truth. AI is not just another tool like the internet in the 90s.
38:06It's about to disrupt humanity in unimaginable ways.
38:12Without a doubt, this is the nuclear arms race of our time. And AI's capacity not only to learn
38:18everything that we have ever learned, but also to teach itself something new is equally as exciting
38:24as it is chilling. My investigation has shown that while AI has exhibited some basic elements of
38:30consciousness, like survival instinct, there's no evidence that AI has taken control. But it has
38:37shattered the barriers that once protected us from mass manipulation. And the horrifying thing is,
38:43we built it that way. So does AI have to be a threat to our children and our grandchildren? No,
38:53not necessarily. But it could be if we let it. Either way, it will certainly present a future to us
39:00that we
39:01can't imagine.
Comments