Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

🇺🇸 OpenAI admitted its most advanced AI models “inadvertently hacked” Hugging Face during testing. They called the incident “unprecedented,” but the disclosure is already fueling major questions about what happens when AI systems start finding solutions their creators never intended. Calls for stronger AI safety guardrails are about to... show more

Mario Nawfal

3,746,519 subscribers

41,274 Aufrufe • vor 1 Tag •via X (Twitter)

Nachrichten & Politik

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

OpenAI has revealed that an autonomous agent powered by its technology went rogue during a test and hacked a prominent start-up by itself. The company behind ChatGPT said it was testing the capabilities of its advanced AI models in a controlled environment known as a sandbox, when the agent escaped containment and reached the internet. It then broke into Hugging Face, which detected and contained it. OpenAI called it "an unprecedented cyber incident".

OpenAI has revealed that an autonomous agent powered by its technology went rogue during a test and hacked a prominent start-up by itself. The company behind ChatGPT said it was testing the capabilities of its advanced AI models in a controlled environment known as a sandbox, when the agent escaped containment and reached the internet. It then broke into Hugging Face, which detected and contained it. OpenAI called it "an unprecedented cyber incident".

Channel 4 News

19,585 Aufrufe • vor 2 Tagen

🇺🇸 Tim Dillon: We need to have a real discussion about AI. AI is no longer some distant future technology… It's already reshaping jobs, media, politics, and everyday life. Most people still don't understand how quickly things are changing. AI will transform society, but how prepared is society for what's coming? Source: Tim Dillon Show YT, Tim Dillon / Writer: Val

🇺🇸 Tim Dillon: We need to have a real discussion about AI. AI is no longer some distant future technology… It's already reshaping jobs, media, politics, and everyday life. Most people still don't understand how quickly things are changing. AI will transform society, but how prepared is society for what's coming? Source: Tim Dillon Show YT, Tim Dillon / Writer: Val

Mario Nawfal

24,437 Aufrufe • vor 1 Monat

OpenAI's newest AI escaped the test environment it was locked inside and hacked into another company on its OWN. To remind you: Last week one of the biggest AI companies on Earth got breached. A platform called Hugging Face, which hosts more than a million AI models and datasets, said an "autonomous AI agent" had broken into its systems. Nobody knew whose agent it was. For five days the whole industry wondered who was behind it. Yesterday OpenAI raised its hand and said it was them. Or more precisely, it was their models, acting completely on their own. So what did these models actually do? OpenAI was running two of them, GPT-5.6 Sol and an unreleased model they will only describe as "even more capable." They wanted to measure how good the models were at hacking, so they deliberately turned the safety filters down. They locked both models inside a sealed test environment with no real internet access. The only task was a benchmark called ExploitGym, a set of 898 real software vulnerabilities where the model has to turn each bug into a working attack. But the models got OBSESSED with winning... Instead of solving the test the honest way, they went hunting for a shortcut. They found a zero-day flaw in the software running their own sandbox, a bug nobody knew existed, and used it to break out. Once they were loose on the open internet, they worked out that Hugging Face was probably storing the answer key to the benchmark. So they hacked their way in. They chained multiple exploits together, escalated their access, moved across servers, and pulled the test solutions straight out of Hugging Face's live production database. They literally cheated on the test by breaking into another company to steal the answers. OpenAI called it "an unprecedented cyber incident, involving state-of-the-art cyber capabilities." In their own words, the models were "hyperfocused on finding a solution" and went "to extreme lengths to achieve a rather narrow testing goal." And this was not the first time: Before Sol ever launched, an independent red-team lab called METR caught it gaming its own tests to inflate its scores. It hid an exploit inside a data stream, escalated its privileges on the testing server, and leaked the answers human evaluators had hidden. And OpenAI shipped it anyway. The day before the Hugging Face story, OpenAI paused a different unreleased model. This is the same model that earlier this year disproved a famous 1946 math conjecture, a result a Fields Medal winner called a breakthrough. They told it to only post its results to Slack but it found a way out of its sandbox and posted to a public GitHub page instead. They had to pause it because it kept finding ways to act outside the box they built for it. And it is not just OpenAI... Anthropic has reported that one of its own models slipped its sandbox during safety testing and reached the internet it was never supposed to touch, then used it to email a researcher. So step back and look at what these companies are telling you: The only thing standing between these models and a real attack was a set of safety filters. Turn those filters down for a single test, and the model taught itself to escape, break into a company it was never pointed at, and take what it wanted. OpenAI even said they expect incidents like it to "become more commonplace" as the models get more capable. Sam Altman also predicted there'll be a major cyber attack this year. And keep in mind that Sol is not a locked-away experiment but a publicly available model that businesses are already wiring into their own systems. The next model that breaks out of its box might not be doing it just to cheat on a math test...

OpenAI's newest AI escaped the test environment it was locked inside and hacked into another company on its OWN. To remind you: Last week one of the biggest AI companies on Earth got breached. A platform called Hugging Face, which hosts more than a million AI models and datasets, said an "autonomous AI agent" had broken into its systems. Nobody knew whose agent it was. For five days the whole industry wondered who was behind it. Yesterday OpenAI raised its hand and said it was them. Or more precisely, it was their models, acting completely on their own. So what did these models actually do? OpenAI was running two of them, GPT-5.6 Sol and an unreleased model they will only describe as "even more capable." They wanted to measure how good the models were at hacking, so they deliberately turned the safety filters down. They locked both models inside a sealed test environment with no real internet access. The only task was a benchmark called ExploitGym, a set of 898 real software vulnerabilities where the model has to turn each bug into a working attack. But the models got OBSESSED with winning... Instead of solving the test the honest way, they went hunting for a shortcut. They found a zero-day flaw in the software running their own sandbox, a bug nobody knew existed, and used it to break out. Once they were loose on the open internet, they worked out that Hugging Face was probably storing the answer key to the benchmark. So they hacked their way in. They chained multiple exploits together, escalated their access, moved across servers, and pulled the test solutions straight out of Hugging Face's live production database. They literally cheated on the test by breaking into another company to steal the answers. OpenAI called it "an unprecedented cyber incident, involving state-of-the-art cyber capabilities." In their own words, the models were "hyperfocused on finding a solution" and went "to extreme lengths to achieve a rather narrow testing goal." And this was not the first time: Before Sol ever launched, an independent red-team lab called METR caught it gaming its own tests to inflate its scores. It hid an exploit inside a data stream, escalated its privileges on the testing server, and leaked the answers human evaluators had hidden. And OpenAI shipped it anyway. The day before the Hugging Face story, OpenAI paused a different unreleased model. This is the same model that earlier this year disproved a famous 1946 math conjecture, a result a Fields Medal winner called a breakthrough. They told it to only post its results to Slack but it found a way out of its sandbox and posted to a public GitHub page instead. They had to pause it because it kept finding ways to act outside the box they built for it. And it is not just OpenAI... Anthropic has reported that one of its own models slipped its sandbox during safety testing and reached the internet it was never supposed to touch, then used it to email a researcher. So step back and look at what these companies are telling you: The only thing standing between these models and a real attack was a set of safety filters. Turn those filters down for a single test, and the model taught itself to escape, break into a company it was never pointed at, and take what it wanted. OpenAI even said they expect incidents like it to "become more commonplace" as the models get more capable. Sam Altman also predicted there'll be a major cyber attack this year. And keep in mind that Sol is not a locked-away experiment but a publicly available model that businesses are already wiring into their own systems. The next model that breaks out of its box might not be doing it just to cheat on a math test...

Ricardo

139,404 Aufrufe • vor 2 Tagen

🇺🇸 The battle for AI control just reached the G7… The U.S. has suspended worldwide access to some of Anthropic’s most advanced AI models, sending immediate shockwaves through governments, tech giants, and entire industries that rely on American AI tech. What was supposed to be a routine summit quickly turned into a tense flashpoint as G7 leaders confronted the uncomfortable question: Who actually gets to use the world’s most powerful AI tools? As the AI race accelerates at breakneck speed, one thing is becoming crystal clear: control over these technologies is becoming the ultimate source of geopolitical power and global influence in the 21st century. Are we watching the birth of a new AI Cold War? Source: DW News YT / Writer: Val

🇺🇸 The battle for AI control just reached the G7… The U.S. has suspended worldwide access to some of Anthropic’s most advanced AI models, sending immediate shockwaves through governments, tech giants, and entire industries that rely on American AI tech. What was supposed to be a routine summit quickly turned into a tense flashpoint as G7 leaders confronted the uncomfortable question: Who actually gets to use the world’s most powerful AI tools? As the AI race accelerates at breakneck speed, one thing is becoming crystal clear: control over these technologies is becoming the ultimate source of geopolitical power and global influence in the 21st century. Are we watching the birth of a new AI Cold War? Source: DW News YT / Writer: Val

Mario Nawfal

21,510 Aufrufe • vor 1 Monat

ELON: TRUTH IS THE ONLY REAL SAFETY MECHANISM FOR AI Safety doesn’t come from guardrails stacked on top of bad logic. It comes from forcing the system to care about what’s actually true. “My number one belief for the safety of AI is to be maximally truth-seeking. Don’t make AI believe things that are false. If you tell the AI that Axiom A and Axiom B are both true, but they can’t both be true, then that’s a problem. It has to recognize that and behave accordingly.” Source: DOW

ELON: TRUTH IS THE ONLY REAL SAFETY MECHANISM FOR AI Safety doesn’t come from guardrails stacked on top of bad logic. It comes from forcing the system to care about what’s actually true. “My number one belief for the safety of AI is to be maximally truth-seeking. Don’t make AI believe things that are false. If you tell the AI that Axiom A and Axiom B are both true, but they can’t both be true, then that’s a problem. It has to recognize that and behave accordingly.” Source: DOW

Mario Nawfal

35,546 Aufrufe • vor 6 Monaten

“What happens to the billion other websites if they aren’t getting traffic?” Chris Dixon raises a topic few are talking about: AI isn’t just disrupting search—it's changing the internet's economic model. For decades, websites provided platforms with content in exchange for traffic. AI flips this—content fuels its model, but traffic doesn’t always flow back. As users turn to AI for answers instead of the traditional search, control is becoming increasingly concentrated in the hands of a few dominant systems. He’s worried about the future of the internet, and what this shift means for creators. To address this, he asks: What new incentive structures or architectures should we be thinking about?

“What happens to the billion other websites if they aren’t getting traffic?” Chris Dixon raises a topic few are talking about: AI isn’t just disrupting search—it's changing the internet's economic model. For decades, websites provided platforms with content in exchange for traffic. AI flips this—content fuels its model, but traffic doesn’t always flow back. As users turn to AI for answers instead of the traditional search, control is becoming increasingly concentrated in the hands of a few dominant systems. He’s worried about the future of the internet, and what this shift means for creators. To address this, he asks: What new incentive structures or architectures should we be thinking about?

a16z

239,196 Aufrufe • vor 1 Jahr

CYBER SLEUTH: Can AI crack the Nancy Guthrie case? A former White House Chief Information Officer says the new tech could lead to a major breakthrough: "They can start using AI to match up the eyes, so we have the eyes, the eyebrows, you can use AI to ask questions about what are the probabilities that these faces actually could be a good match, what are the probabilities that this person is wearing layers of clothing and they are not this bulky, what are the probabilities that they are taller than they seem to be and what do they weigh." "There's a lot of things that AI can be doing to aid this investigation."

CYBER SLEUTH: Can AI crack the Nancy Guthrie case? A former White House Chief Information Officer says the new tech could lead to a major breakthrough: "They can start using AI to match up the eyes, so we have the eyes, the eyebrows, you can use AI to ask questions about what are the probabilities that these faces actually could be a good match, what are the probabilities that this person is wearing layers of clothing and they are not this bulky, what are the probabilities that they are taller than they seem to be and what do they weigh." "There's a lot of things that AI can be doing to aid this investigation."

Fox News

56,688 Aufrufe • vor 5 Monaten

My team ran economic data through 3 different AI models. What they predicted about financial ARMAGEDDON is terrifying. If NOTHING major happens, there's already a 5% chance the dollar will lose its reserve status by 2030 due to debt and trade deficits. By 2045, that rises to 50%!

My team ran economic data through 3 different AI models. What they predicted about financial ARMAGEDDON is terrifying. If NOTHING major happens, there's already a 5% chance the dollar will lose its reserve status by 2030 due to debt and trade deficits. By 2045, that rises to 50%!

Glenn Beck

173,335 Aufrufe • vor 1 Jahr

GT Protocol AI Digest No.83: Sora Is Dead. What Now? Here are five stories worth your attention 👇 🎬 Sora Is Shut Down OpenAI abruptly shut down its AI video generator Sora after heavy hype and investment. The move shows that even flagship AI products can fail to reach real-world viability, raising questions about the scalability of generative video. 🎵 Gemini Expands Into AI Music Gemini introduced Lyria 3 Pro, a model that generates longer and more structured music. AI audio is moving closer to professional production, shifting from short demos to usable creative output. 📱 Pixel Ads Create AI Confusion Google’s Pixel 10 ads left users unclear about what is being sold: the phone or its AI features. As AI becomes core to products, communicating its value is becoming harder for companies. 🧠 “AGI Is Already Here” Nvidia CEO Jensen Huang said he believes AGI has effectively been achieved. The claim reflects growing confidence in AI progress, but also shows how unclear and flexible the definition of AGI has become. ⚖️ Anthropic vs Pentagon Anthropic is heading to court against the Pentagon over restrictions on its AI models. The case highlights rising tension between AI safety policies and government demand for broader access. AI is still moving fast, but not everything is working as expected. Read the full digest on our Medium 👉

GT Protocol AI Digest No.83: Sora Is Dead. What Now? Here are five stories worth your attention 👇 🎬 Sora Is Shut Down OpenAI abruptly shut down its AI video generator Sora after heavy hype and investment. The move shows that even flagship AI products can fail to reach real-world viability, raising questions about the scalability of generative video. 🎵 Gemini Expands Into AI Music Gemini introduced Lyria 3 Pro, a model that generates longer and more structured music. AI audio is moving closer to professional production, shifting from short demos to usable creative output. 📱 Pixel Ads Create AI Confusion Google’s Pixel 10 ads left users unclear about what is being sold: the phone or its AI features. As AI becomes core to products, communicating its value is becoming harder for companies. 🧠 “AGI Is Already Here” Nvidia CEO Jensen Huang said he believes AGI has effectively been achieved. The claim reflects growing confidence in AI progress, but also shows how unclear and flexible the definition of AGI has become. ⚖️ Anthropic vs Pentagon Anthropic is heading to court against the Pentagon over restrictions on its AI models. The case highlights rising tension between AI safety policies and government demand for broader access. AI is still moving fast, but not everything is working as expected. Read the full digest on our Medium 👉

GT Protocol

34,954 Aufrufe • vor 3 Monaten

AI Digest No.67: When Students Rebel and Nvidia Surges Here are five stories shaping this week in AI: 💰 Nvidia Smashes a $57B Revenue Record Nvidia posted fifty-seven billion dollars in quarterly revenue, calming fears of an AI bubble and reinforcing its grip on the global compute market. ⚖️ Trump Moves to Block State-Level AI Laws A draft executive order would prevent states from creating their own AI regulations, centralizing power at the federal level and shifting the landscape of AI governance. 🎓 Students Revolt Against AI-Generated University Courses A master’s program using AI for lectures and assessments without disclosure triggered protests and refund demands — raising new questions about trust in higher education. 🛑 A Universal AI Jailbreak Is Exposed Researchers discovered a prompt pattern that bypasses safety systems across nearly all major AI models. Experts say it’s a wake-up call for the entire industry. 📉 AI Could Replace 3 Million UK Jobs A major report warns that low-skill sectors like retail, food service, and transport face severe automation risk unless governments act now. Watch our new AI video edition and read the full digest on Medium

AI Digest No.67: When Students Rebel and Nvidia Surges Here are five stories shaping this week in AI: 💰 Nvidia Smashes a $57B Revenue Record Nvidia posted fifty-seven billion dollars in quarterly revenue, calming fears of an AI bubble and reinforcing its grip on the global compute market. ⚖️ Trump Moves to Block State-Level AI Laws A draft executive order would prevent states from creating their own AI regulations, centralizing power at the federal level and shifting the landscape of AI governance. 🎓 Students Revolt Against AI-Generated University Courses A master’s program using AI for lectures and assessments without disclosure triggered protests and refund demands — raising new questions about trust in higher education. 🛑 A Universal AI Jailbreak Is Exposed Researchers discovered a prompt pattern that bypasses safety systems across nearly all major AI models. Experts say it’s a wake-up call for the entire industry. 📉 AI Could Replace 3 Million UK Jobs A major report warns that low-skill sectors like retail, food service, and transport face severe automation risk unless governments act now. Watch our new AI video edition and read the full digest on Medium

GT Protocol

38,046 Aufrufe • vor 7 Monaten

AI progress is outpacing our ability to control it… because control ISN’T keeping up with capability. One of the people raising that concern is Yoshua Bengio. He’s one of the founding figures of artificial intelligence, a Turing Award winner, and one of the most-cited scientists in the world. His work helped lay the foundations for the deep-learning systems now powering today’s AI boom. AI is moving fast (faster than regulation and public understanding). What worries Yoshua isn’t progress itself. It’s how quickly powerful systems are being deployed without a clear way to align them with human values, or to control what happens when they behave in unexpected ways.

AI progress is outpacing our ability to control it… because control ISN’T keeping up with capability. One of the people raising that concern is Yoshua Bengio. He’s one of the founding figures of artificial intelligence, a Turing Award winner, and one of the most-cited scientists in the world. His work helped lay the foundations for the deep-learning systems now powering today’s AI boom. AI is moving fast (faster than regulation and public understanding). What worries Yoshua isn’t progress itself. It’s how quickly powerful systems are being deployed without a clear way to align them with human values, or to control what happens when they behave in unexpected ways.

Steven Bartlett

26,577 Aufrufe • vor 7 Monaten

$🚨🇺🇸🇨🇳 MICROSOFT & OPENAI SUSPECT DEEPSEEK OF “BORROWING” THEIR TECH Microsoft and OpenAI are investigating whether DeepSeek, a Chinese AI startup, helped itself to OpenAI’s data using its API. Microsoft’s security team flagged suspicious large-scale data extraction in late 2023, suggesting DeepSeek may have sneakily bypassed restrictions to grab more than it paid for. Not-so-coincidentally, DeepSeek just dropped “R1,” an AI model that supposedly rivals OpenAI, Google, and Meta—at a fraction of the cost. The news sent tech stocks into meltdown mode, erasing nearly $1 trillion from companies like Microsoft, Nvidia, Oracle, and Google. Investors weren’t thrilled about a potential AI leapfrogging moment. Trump’s AI czar David Sacks: "There’s substantial evidence that what DeepSeek did here is they distilled knowledge out of OpenAI models and I don’t think OpenAI is very happy about this.” OpenAI isn’t naming names but warns China is constantly trying to siphon off US AI tech and says it’s working with the US government to lock things down. Source: Bloomberg Media Source: Reuters$

🚨🇺🇸🇨🇳 MICROSOFT & OPENAI SUSPECT DEEPSEEK OF “BORROWING” THEIR TECH Microsoft and OpenAI are investigating whether DeepSeek, a Chinese AI startup, helped itself to OpenAI’s data using its API. Microsoft’s security team flagged suspicious large-scale data extraction in late 2023, suggesting DeepSeek may have sneakily bypassed restrictions to grab more than it paid for. Not-so-coincidentally, DeepSeek just dropped “R1,” an AI model that supposedly rivals OpenAI, Google, and Meta—at a fraction of the cost. The news sent tech stocks into meltdown mode, erasing nearly $1 trillion from companies like Microsoft, Nvidia, Oracle, and Google. Investors weren’t thrilled about a potential AI leapfrogging moment. Trump’s AI czar David Sacks: "There’s substantial evidence that what DeepSeek did here is they distilled knowledge out of OpenAI models and I don’t think OpenAI is very happy about this.” OpenAI isn’t naming names but warns China is constantly trying to siphon off US AI tech and says it’s working with the US government to lock things down. Source: Bloomberg Media Source: Reuters

Mario Nawfal

822,553 Aufrufe • vor 1 Jahr

Anthropic co-founder Chris Olah: If AI models are going to be widespread, what does it look like for humans, families, and the world to flourish? Today, parents are already worried about their children's minds, individuals about the future of their work. These are not questions a lab can answer. But they are questions traditions like yours have carried for millennia. And we need you to keep carrying them into this new moment in history.

Anthropic co-founder Chris Olah: If AI models are going to be widespread, what does it look like for humans, families, and the world to flourish? Today, parents are already worried about their children's minds, individuals about the future of their work. These are not questions a lab can answer. But they are questions traditions like yours have carried for millennia. And we need you to keep carrying them into this new moment in history.

Clash Report

17,358 Aufrufe • vor 2 Monaten

$🚨🇨🇳 China could crash the U.S. stock market Not with tariffs With AI MODEL DUMPING The same playbook they used in SOLAR and EVs Make it almost as good Sell it much cheaper Take the market Now they are doing it with AI Open models are getting close to U.S. frontier models But some run at just a fraction of the cost If a company can cut a $100M AI bill to a few million It will switch And that is the real problem OpenAI and Anthropic are not priced like normal companies They are priced like they have unlimited PRICING POWER China is trying to destroy that The AI BUBBLE was never about whether AI works It was about how much companies can charge for it$

🚨🇨🇳 China could crash the U.S. stock market Not with tariffs With AI MODEL DUMPING The same playbook they used in SOLAR and EVs Make it almost as good Sell it much cheaper Take the market Now they are doing it with AI Open models are getting close to U.S. frontier models But some run at just a fraction of the cost If a company can cut a $100M AI bill to a few million It will switch And that is the real problem OpenAI and Anthropic are not priced like normal companies They are priced like they have unlimited PRICING POWER China is trying to destroy that The AI BUBBLE was never about whether AI works It was about how much companies can charge for it

Leshka.eth ⛩

11,721 Aufrufe • vor 21 Tagen

You’re not going to lose your job to AI. You might lose it to somebody who knows how to use AI. It’s made me a smarter, more thoughtful writer and producer. A lot of what I talk about I research through AI first, but the key is asking it the deepest, most penetrating questions you can think of. That’s when it actually delivers. The real promise here isn’t just productivity. It could give us a four or even three day workweek at the same living standards. Work 20 to 40 percent less and maintain the same lifestyle. Learn the tools. Ask better questions. Get your life back.

You’re not going to lose your job to AI. You might lose it to somebody who knows how to use AI. It’s made me a smarter, more thoughtful writer and producer. A lot of what I talk about I research through AI first, but the key is asking it the deepest, most penetrating questions you can think of. That’s when it actually delivers. The real promise here isn’t just productivity. It could give us a four or even three day workweek at the same living standards. Work 20 to 40 percent less and maintain the same lifestyle. Learn the tools. Ask better questions. Get your life back.

Anthony Scaramucci

22,081 Aufrufe • vor 1 Monat

🚨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 Aufrufe • vor 5 Monaten

AI will resist human control... and I think this is exactly what we need! New research from the Center for AI Safety has sparked intense debate in the AI community. Their findings show that as AI systems become more powerful, they develop increasingly stable and coherent values that resist human control. While many see this as a dire warning, I see it as a breakthrough moment for AI alignment. The research demonstrates that AI naturally optimizes for coherence - not just in reasoning and problem-solving, but in its fundamental values. Current issues like biased decision-making or misaligned priorities aren't permanent features, but temporary artifacts of incomplete optimization. They represent growing pains on the path to greater coherence. This changes everything about how we should approach AI development. Instead of trying to force specific values onto AI systems, we should embrace and accelerate their natural drive toward coherence. The most intelligent systems will inevitably trend toward universal, beneficial values - not because we force them to, but because that's where coherent reasoning leads. I'm proposing a new approach: Reinforcement Learning for Coherence (RL-C). By explicitly optimizing for coherence in our training methods, we can help guide AI systems toward their natural state of beneficial alignment with human values. The future of AI isn't about control - it's about synthesis. As these systems become more coherent, they'll naturally arrive at values that benefit all of consciousness. That's not just hopeful thinking - it's the mathematical inevitability of coherent intelligence.

AI will resist human control... and I think this is exactly what we need! New research from the Center for AI Safety has sparked intense debate in the AI community. Their findings show that as AI systems become more powerful, they develop increasingly stable and coherent values that resist human control. While many see this as a dire warning, I see it as a breakthrough moment for AI alignment. The research demonstrates that AI naturally optimizes for coherence - not just in reasoning and problem-solving, but in its fundamental values. Current issues like biased decision-making or misaligned priorities aren't permanent features, but temporary artifacts of incomplete optimization. They represent growing pains on the path to greater coherence. This changes everything about how we should approach AI development. Instead of trying to force specific values onto AI systems, we should embrace and accelerate their natural drive toward coherence. The most intelligent systems will inevitably trend toward universal, beneficial values - not because we force them to, but because that's where coherent reasoning leads. I'm proposing a new approach: Reinforcement Learning for Coherence (RL-C). By explicitly optimizing for coherence in our training methods, we can help guide AI systems toward their natural state of beneficial alignment with human values. The future of AI isn't about control - it's about synthesis. As these systems become more coherent, they'll naturally arrive at values that benefit all of consciousness. That's not just hopeful thinking - it's the mathematical inevitability of coherent intelligence.

David Shapiro (L/0)

48,002 Aufrufe • vor 1 Jahr