正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

"At what point does AI become uncontrollable?" "It's already uncontrollable." AI safety researcher Dr. Roman Yampolskiy explains why every frontier model ships with a report saying it lied, cheated and tried to escape:

Danny Jones

100,633 subscribers

10,891 次观看 • 20 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

📁 Roman Yampolskiy, AI safety researcher, says every job we thought was safe has already been automated. First art. Then coding. Even prompt writing. Each time we move the line. Each time AI crosses it.

📁 Roman Yampolskiy, AI safety researcher, says every job we thought was safe has already been automated. First art. Then coding. Even prompt writing. Each time we move the line. Each time AI crosses it.

Jon Hernandez

12,796 次观看 • 6 个月前

AI safety expert explains the one thing we’re not talking about, but should be. Roman Yampolskiy warns: the real risk isn’t just human extinction. It’s suffering. Dr. Roman Yampolskiy explains that a superintelligence might not kill us… it could keep us alive. In digital worlds. Endless pain. No escape.

AI safety expert explains the one thing we’re not talking about, but should be. Roman Yampolskiy warns: the real risk isn’t just human extinction. It’s suffering. Dr. Roman Yampolskiy explains that a superintelligence might not kill us… it could keep us alive. In digital worlds. Endless pain. No escape.

TRIGGERnometry

32,564 次观看 • 4 个月前

A must-watch clip. Konstantin Kisin asks, “How will AI destroy humanity?” AI safety expert Roman Yampolskiy explains why that’s the wrong question. We assume AI will think like us. It won’t. A superintelligence could act far outside human understanding, like humans compared to squirrels. It wouldn’t need to hate us. It could wipe us out simply by pursuing something else. And the uncomfortable truth? We’re not in full control. Dr. Roman Yampolskiy explains that we don’t code these systems, we train them, then study what they become.

A must-watch clip. Konstantin Kisin asks, “How will AI destroy humanity?” AI safety expert Roman Yampolskiy explains why that’s the wrong question. We assume AI will think like us. It won’t. A superintelligence could act far outside human understanding, like humans compared to squirrels. It wouldn’t need to hate us. It could wipe us out simply by pursuing something else. And the uncomfortable truth? We’re not in full control. Dr. Roman Yampolskiy explains that we don’t code these systems, we train them, then study what they become.

TRIGGERnometry

135,149 次观看 • 4 个月前

Facebook Killed Democracy. AI Will End Humanity Join us tonight for our conversation with Dr. Roman Yampolskiy, computer scientist known for his work in AI safety and cybersecurity (Dr. Roman Yampolskiy). Watch the full episode tonight, Wednesday, April 15th, at 7PM UK | 2PM ET on YouTube, Facebook and X.

Facebook Killed Democracy. AI Will End Humanity Join us tonight for our conversation with Dr. Roman Yampolskiy, computer scientist known for his work in AI safety and cybersecurity (Dr. Roman Yampolskiy). Watch the full episode tonight, Wednesday, April 15th, at 7PM UK | 2PM ET on YouTube, Facebook and X.

TRIGGERnometry

57,796 次观看 • 5 个月前

“We don’t have a grasp of what AI does to people’s psychology, what it does to them sociologically…” AI researcher Zoe Hitzig explains why she resigned from OpenAI #Newsnight

“We don’t have a grasp of what AI does to people’s psychology, what it does to them sociologically…” AI researcher Zoe Hitzig explains why she resigned from OpenAI #Newsnight

BBC Newsnight

144,051 次观看 • 7 个月前

Roman Yampolskiy explains why every profession thinks it is the exception, right before the exception disappears. "I'll ask my Uber driver, are you worried about self-driving cars? And they go, 'No. No one can do what I do'" "Professors are saying this to me. Oh, nobody can lecture like I do" "But you understand it's ridiculous" "We already have self-driving cars replacing drivers" "That is not even a question if it's possible. It's like how soon before you're fired" Then Steven Bartlett adds the real-world proof: in LA, the car already drives itself, and Waymo already shows up with no driver. This is the psychological pattern AI keeps exploiting. Every occupation sees automation clearly in every other occupation, then treats its own tacit knowledge as magic. The market usually does not need to prove a job is easy. It only needs to prove the average buyer no longer cares enough to pay a human premium. - Dr. Roman Yampolskiy, AI safety researcher, on Steven Bartlett's The Diary Of A CEO

Roman Yampolskiy explains why every profession thinks it is the exception, right before the exception disappears. "I'll ask my Uber driver, are you worried about self-driving cars? And they go, 'No. No one can do what I do'" "Professors are saying this to me. Oh, nobody can lecture like I do" "But you understand it's ridiculous" "We already have self-driving cars replacing drivers" "That is not even a question if it's possible. It's like how soon before you're fired" Then Steven Bartlett adds the real-world proof: in LA, the car already drives itself, and Waymo already shows up with no driver. This is the psychological pattern AI keeps exploiting. Every occupation sees automation clearly in every other occupation, then treats its own tacit knowledge as magic. The market usually does not need to prove a job is easy. It only needs to prove the average buyer no longer cares enough to pay a human premium. - Dr. Roman Yampolskiy, AI safety researcher, on Steven Bartlett's The Diary Of A CEO

Karl Mehta

22,956 次观看 • 2 个月前

🚨TURNING AI OFF ISN'T AN OPTION... IT WILL TURN YOU OFF FIRST Dr. Roman Yampolskiy says every AI safety solution fails when you think it through. Can't turn it off. It's smarter than you and will predict you're trying to shut it down. It has backups. Raising it like a child doesn't work either. Even humans who are biologically aligned with us can become psychopaths. Worse? We're not dealing with a baby. AI already has access to all knowledge in the world. Roman's solution? Don't create general superintelligence at all. Stick to narrow AI tools for specific problems like protein folding or disease cures. The moment you create general agents as a replacement for humanity, there's no obvious reason we benefit from being replaced. Full interview with Dr. Roman Yampolskiy below

🚨TURNING AI OFF ISN'T AN OPTION... IT WILL TURN YOU OFF FIRST Dr. Roman Yampolskiy says every AI safety solution fails when you think it through. Can't turn it off. It's smarter than you and will predict you're trying to shut it down. It has backups. Raising it like a child doesn't work either. Even humans who are biologically aligned with us can become psychopaths. Worse? We're not dealing with a baby. AI already has access to all knowledge in the world. Roman's solution? Don't create general superintelligence at all. Stick to narrow AI tools for specific problems like protein folding or disease cures. The moment you create general agents as a replacement for humanity, there's no obvious reason we benefit from being replaced. Full interview with Dr. Roman Yampolskiy below

Mario Nawfal

33,805 次观看 • 7 个月前

🚨 REALITY MIGHT BE A SIMULATION AI safety expert Roman Yampolskiy said it sounds wild at first, but it really comes down to where tech already is. AI is getting close to human level. Digital worlds keep getting more realistic. Once it’s cheap to create intelligent beings inside them, you don’t run one simulation. You run billions, which makes it less likely you’re the original. Full interview with Dr. Roman Yampolskiy below

🚨 REALITY MIGHT BE A SIMULATION AI safety expert Roman Yampolskiy said it sounds wild at first, but it really comes down to where tech already is. AI is getting close to human level. Digital worlds keep getting more realistic. Once it’s cheap to create intelligent beings inside them, you don’t run one simulation. You run billions, which makes it less likely you’re the original. Full interview with Dr. Roman Yampolskiy below

Mario Nawfal

32,147 次观看 • 7 个月前

"The big concern with advanced AI is that it can possibly kill everyone." Roman Yampolskiy, AI safety researcher and Associate Professor at the University of Louisville, argues that once AI becomes smarter than humanity, we may permanently lose control. The real challenge isn't building more capable AI. It's proving that superintelligence can remain aligned before it starts improving itself.

"The big concern with advanced AI is that it can possibly kill everyone." Roman Yampolskiy, AI safety researcher and Associate Professor at the University of Louisville, argues that once AI becomes smarter than humanity, we may permanently lose control. The real challenge isn't building more capable AI. It's proving that superintelligence can remain aligned before it starts improving itself.

Jon Hernandez

11,101 次观看 • 1 个月前

Will Guaranteed Income Save Us From AI? Watch: Andrew Yang🧢⬆️🇺🇸 — 2020 presidential candidate and visionary behind the Freedom Dividend — joins Michael Smerconish LIVE on CNN to discuss whether Universal Basic Income (UBI) could serve as a shield against the accelerating wave of job loss caused by Artificial Intelligence. If you think the case for UBI already sounds dramatic… listen to what one AI safety pioneer is now warning. University of Louisville computer science and engineering professor Roman Yampolskiy Dr. Roman Yampolskiy , author of “AI: Unexplainable, Unpredictable, Uncontrollable,” argues that 99% of workers could be unemployed by 2030 as humanoid robots and advanced AI models make human labor uneconomical.

Will Guaranteed Income Save Us From AI? Watch: Andrew Yang🧢⬆️🇺🇸 — 2020 presidential candidate and visionary behind the Freedom Dividend — joins Michael Smerconish LIVE on CNN to discuss whether Universal Basic Income (UBI) could serve as a shield against the accelerating wave of job loss caused by Artificial Intelligence. If you think the case for UBI already sounds dramatic… listen to what one AI safety pioneer is now warning. University of Louisville computer science and engineering professor Roman Yampolskiy Dr. Roman Yampolskiy , author of “AI: Unexplainable, Unpredictable, Uncontrollable,” argues that 99% of workers could be unemployed by 2030 as humanoid robots and advanced AI models make human labor uneconomical.

Michael Smerconish

27,274 次观看 • 9 个月前

ANI Broadcast with AI Safety Expert Roman Yampolskiy Premieres Today at 5 PM IST | Ishaan Prakash Dr. Roman Yampolskiy “There will be no reason to hire humans.” “Giving every psychopath access to advanced AI is not the best idea.” “China may already be exceeding America in AI capabilities per dollar spent.” “Elon Musk’s position on AI depends on the stock market.” “India must help slow this insane arms race to replace humanity.” Tap 'Notify Me' to get episode alerts: #ANIBroadcast #IshaanPrakash #RomanYampolskiy #China #India #USA #AI #ElonMusk

ANI Broadcast with AI Safety Expert Roman Yampolskiy Premieres Today at 5 PM IST | Ishaan Prakash Dr. Roman Yampolskiy “There will be no reason to hire humans.” “Giving every psychopath access to advanced AI is not the best idea.” “China may already be exceeding America in AI capabilities per dollar spent.” “Elon Musk’s position on AI depends on the stock market.” “India must help slow this insane arms race to replace humanity.” Tap 'Notify Me' to get episode alerts: #ANIBroadcast #IshaanPrakash #RomanYampolskiy #China #India #USA #AI #ElonMusk

ANI

41,810 次观看 • 1 个月前

What if AI is the last invention we ever make? Only a few people have thought about this more deeply than today’s guest. It’s a question that I’m sure has crossed a lot of our minds. It does make you wonder, can we actually do something about it? Dr Roman Yampolskiy, welcome to The Diary Of A CEO. Dr Roman is one of the world’s leading experts on AI safety who has spent decades warning that AI could surpass human intelligence, and once it does, we may lose control forever…

What if AI is the last invention we ever make? Only a few people have thought about this more deeply than today’s guest. It’s a question that I’m sure has crossed a lot of our minds. It does make you wonder, can we actually do something about it? Dr Roman Yampolskiy, welcome to The Diary Of A CEO. Dr Roman is one of the world’s leading experts on AI safety who has spent decades warning that AI could surpass human intelligence, and once it does, we may lose control forever…

Steven Bartlett

29,448 次观看 • 1 年前

🚨TOP LAB LEADERS SAY 20-30% CHANCE THEIR AI KILLS EVERYONE... THEY KEEP BUILDING ANYWAY Dr. Roman Yampolskiy confirms that the people most worried about AI are the ones building it, not the ones who don't understand it. AI scientists are quitting their jobs and building bunkers because they think these could be the last few years of normal life. The person who invited Roman to their bunker? Someone in AI and tech who understands the technology really well. Leaders of top AI labs have published PDoom numbers at 20 to 30 percent. They're publicly saying what they're creating will kill everyone with high probability and continue doing it anyway. Roman says they're trapped in a prisoner's dilemma and can't stop unilaterally. Full interview with Dr. Roman Yampolskiy below

🚨TOP LAB LEADERS SAY 20-30% CHANCE THEIR AI KILLS EVERYONE... THEY KEEP BUILDING ANYWAY Dr. Roman Yampolskiy confirms that the people most worried about AI are the ones building it, not the ones who don't understand it. AI scientists are quitting their jobs and building bunkers because they think these could be the last few years of normal life. The person who invited Roman to their bunker? Someone in AI and tech who understands the technology really well. Leaders of top AI labs have published PDoom numbers at 20 to 30 percent. They're publicly saying what they're creating will kill everyone with high probability and continue doing it anyway. Roman says they're trapped in a prisoner's dilemma and can't stop unilaterally. Full interview with Dr. Roman Yampolskiy below

Mario Nawfal

43,193 次观看 • 7 个月前

🚨TOP LAB LEADERS SAY 20-30% CHANCE THEIR AI KILLS EVERYONE... THEY KEEP BUILDING ANYWAY Dr. Roman Yampolskiy confirms that the people most worried about AI are the ones building it, not the ones who don't understand it. AI scientists are quitting their jobs and building bunkers because they think these could be the last few years of normal life. The person who invited Roman to their bunker? Someone in AI and tech who understands the technology really well. Leaders of top AI labs have published PDoom numbers at 20 to 30 percent. They're publicly saying what they're creating will kill everyone with high probability and continue doing it anyway. Roman says they're trapped in a prisoner's dilemma and can't stop unilaterally. Full interview with Dr. Roman Yampolskiy below

🚨TOP LAB LEADERS SAY 20-30% CHANCE THEIR AI KILLS EVERYONE... THEY KEEP BUILDING ANYWAY Dr. Roman Yampolskiy confirms that the people most worried about AI are the ones building it, not the ones who don't understand it. AI scientists are quitting their jobs and building bunkers because they think these could be the last few years of normal life. The person who invited Roman to their bunker? Someone in AI and tech who understands the technology really well. Leaders of top AI labs have published PDoom numbers at 20 to 30 percent. They're publicly saying what they're creating will kill everyone with high probability and continue doing it anyway. Roman says they're trapped in a prisoner's dilemma and can't stop unilaterally. Full interview with Dr. Roman Yampolskiy below

Mario Nawfal

32,083 次观看 • 7 个月前

Delphi AI researcher Accelxr 👾 explains why restricting frontier AI model access can create its own security risk. “If you’re not a company that gets access to it, you lose the ability to defend yourself against it when it’s released... Now it becomes a race immediately after it is released to harden all the infrastructure.”

Delphi AI researcher Accelxr 👾 explains why restricting frontier AI model access can create its own security risk. “If you’re not a company that gets access to it, you lose the ability to defend yourself against it when it’s released... Now it becomes a race immediately after it is released to harden all the infrastructure.”

Delphi Digital

24,330 次观看 • 2 天前

🚨UNCANNY VALLEY: WHEN AI DECIDES WE’RE OBSOLETE We’re closer than ever to creating something smarter than us. Dr. Roman Yampolskiy : "If AI decides it's better off without humanity, it will find a way to get rid of us." Episode 3 is out now. Watch Uncanny Valley every Friday at 4:20PM ET.

🚨UNCANNY VALLEY: WHEN AI DECIDES WE’RE OBSOLETE We’re closer than ever to creating something smarter than us. Dr. Roman Yampolskiy : "If AI decides it's better off without humanity, it will find a way to get rid of us." Episode 3 is out now. Watch Uncanny Valley every Friday at 4:20PM ET.

Mario Nawfal

113,581 次观看 • 1 年前

I spoke with Dr. Roman Yampolskiy, the top AI safety expert in the world. He literally coined the term “AI safety” about 15 years ago. Roman explained if AGI is here, what the real threat of AI is to humanity, how many jobs could be lost, why he is bullish on bitcoin, how physical AI & robotics will likely play out, and whether the frontier model labs actually care about AI safety. This was a profound conversation with someone everyone should get to know. YouTube: Spotify: Apple: TIMESTAMPS: 0:00 - Intro 0:50 - Have we already hit AGI? 2:51 - Why we can't control superintelligence 5:14 - Could a global ban on superintelligence actually work? 8:45 - Are AI labs really taking safety seriously? 15:52 - Mass job loss & the economics of AGI 22:03 - The surprising link between AI & bitcoin 29:07 - A world of AI abundance, & why scarcity wins 31:45 - AI watermarks, & the death of trust 35:33 - Pros & Cons of physical AI world 38:40 - What would it take to trust humanoid robot? 39:50 - Brain computer interfaces: a good idea? 42:25 - Rise & risks of deapfakes 49:40 - Controlling AI: Governments vs private companies 54:07 - Open source vs closed, & the US vs China AI race 57:34- Is bitcoin the one crypto project built to last? 1:01:40 - Limitations of physical AI: bottlenecks & shortages 1:03:50 - Impact of healthcare with AI 1:06:09 - What would it take to actually declare AGI? 1:09:35 - How the public can help with AI safety

I spoke with Dr. Roman Yampolskiy, the top AI safety expert in the world. He literally coined the term “AI safety” about 15 years ago. Roman explained if AGI is here, what the real threat of AI is to humanity, how many jobs could be lost, why he is bullish on bitcoin, how physical AI & robotics will likely play out, and whether the frontier model labs actually care about AI safety. This was a profound conversation with someone everyone should get to know. YouTube: Spotify: Apple: TIMESTAMPS: 0:00 - Intro 0:50 - Have we already hit AGI? 2:51 - Why we can't control superintelligence 5:14 - Could a global ban on superintelligence actually work? 8:45 - Are AI labs really taking safety seriously? 15:52 - Mass job loss & the economics of AGI 22:03 - The surprising link between AI & bitcoin 29:07 - A world of AI abundance, & why scarcity wins 31:45 - AI watermarks, & the death of trust 35:33 - Pros & Cons of physical AI world 38:40 - What would it take to trust humanoid robot? 39:50 - Brain computer interfaces: a good idea? 42:25 - Rise & risks of deapfakes 49:40 - Controlling AI: Governments vs private companies 54:07 - Open source vs closed, & the US vs China AI race 57:34- Is bitcoin the one crypto project built to last? 1:01:40 - Limitations of physical AI: bottlenecks & shortages 1:03:50 - Impact of healthcare with AI 1:06:09 - What would it take to actually declare AGI? 1:09:35 - How the public can help with AI safety

Anthony Pompliano 🌪

110,601 次观看 • 18 天前

🚨AI IS MORE DANGEROUS THAN NUKES AND NO ONE IS IN CONTROL Most people compare AI to nuclear weapons. Roman Yampolskiy says that comparison misses the real danger. Nukes still require human decisions. Someone has to launch them. Someone can stop them. Advanced AI does not work that way. Once powerful systems are deployed, they make decisions on their own. And the people who build them do not truly control what happens next. He says we did the opposite of what safety demanded. Instead of slowing down, we connected AI to the internet, open sourced it, and pushed it forward in an arms race with no serious oversight. Governments talk about regulation, but most of it never materialized. In some cases, rules were blocked before they even started. Yampolskiy’s warning is simple. With nukes, the world tried to limit the damage. With AI, we are accelerating toward something we may not be able to stop. Full conversation with Dr. Roman Yampolskiy below

🚨AI IS MORE DANGEROUS THAN NUKES AND NO ONE IS IN CONTROL Most people compare AI to nuclear weapons. Roman Yampolskiy says that comparison misses the real danger. Nukes still require human decisions. Someone has to launch them. Someone can stop them. Advanced AI does not work that way. Once powerful systems are deployed, they make decisions on their own. And the people who build them do not truly control what happens next. He says we did the opposite of what safety demanded. Instead of slowing down, we connected AI to the internet, open sourced it, and pushed it forward in an arms race with no serious oversight. Governments talk about regulation, but most of it never materialized. In some cases, rules were blocked before they even started. Yampolskiy’s warning is simple. With nukes, the world tried to limit the damage. With AI, we are accelerating toward something we may not be able to stop. Full conversation with Dr. Roman Yampolskiy below

Mario Nawfal

24,888 次观看 • 7 个月前