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。

Anthropic, the tech company that recently clashed with the Pentagon, says it has developed a new AI model that is too powerful to be released to the public.

Channel 4 News

2,302,750 subscribers

103,792 次观看 • 3 个月前 •via X (Twitter)

新闻政治科学技术

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

Anthropic has created an AI model that “can get into almost any software in the world,” former AI company founder and CEO Matt Shumer says — but the company is not releasing it to the public: “I have never seen something like this — a company that is about to IPO and having such a moral compass.” Shumer shares what makes the AI model so powerful and potentially dangerous.

Anthropic has created an AI model that “can get into almost any software in the world,” former AI company founder and CEO Matt Shumer says — but the company is not releasing it to the public: “I have never seen something like this — a company that is about to IPO and having such a moral compass.” Shumer shares what makes the AI model so powerful and potentially dangerous.

CBS Mornings

98,892 次观看 • 3 个月前

Anthropic's newest AI model, called Mythos, is so powerful, the company says, that it could cause widespread disruption if it's released to the public. For now, Anthropic is giving more than 40 tech companies, including some rivals, access to Mythos to test it and identify vulnerabilities across systems. But that move is raising concerns. Watch Geoff Bennett's conversation with Gerrit De Vynck 🦭.

Anthropic's newest AI model, called Mythos, is so powerful, the company says, that it could cause widespread disruption if it's released to the public. For now, Anthropic is giving more than 40 tech companies, including some rivals, access to Mythos to test it and identify vulnerabilities across systems. But that move is raising concerns. Watch Geoff Bennett's conversation with Gerrit De Vynck 🦭.

PBS News

55,148 次观看 • 3 个月前

In an exclusive interview with CBS News' Jo Ling Kent, Anthropic CEO Dario Amodei said that the Pentagon’s decision to designate the AI company a supply chain risk is “retaliatory and punitive.” The Pentagon made the designation, which restricts military contractors from doing business with Anthropic, after the company refused to give the military unfettered access to its AI model.

In an exclusive interview with CBS News' Jo Ling Kent, Anthropic CEO Dario Amodei said that the Pentagon’s decision to designate the AI company a supply chain risk is “retaliatory and punitive.” The Pentagon made the designation, which restricts military contractors from doing business with Anthropic, after the company refused to give the military unfettered access to its AI model.

CBS News

1,213,350 次观看 • 5 个月前

😱Hard to believe this is AI-generated! China’s tech company ByteDance has released its new Seedance2.0 AI video model. This is gonna reshape the film industry!

😱Hard to believe this is AI-generated! China’s tech company ByteDance has released its new Seedance2.0 AI video model. This is gonna reshape the film industry!

Li Zexin 李泽欣

104,127 次观看 • 5 个月前

The Pentagon has effectively blacklisted AI firm Anthropic, after the company refused to agree to the government's demand for unrestricted use of its latest model. "The big question now is, who's going to step up after them?" says technology journalist Jacob Ward.

The Pentagon has effectively blacklisted AI firm Anthropic, after the company refused to agree to the government's demand for unrestricted use of its latest model. "The big question now is, who's going to step up after them?" says technology journalist Jacob Ward.

Anderson Cooper 360°

39,399 次观看 • 5 个月前

Anthropic has withdrawn its powerful Fable 5 model after the US government ordered the company to revoke access to all foreign nationals Anthropic itself described Mythos, Fable 5's base model, as "too powerful". Rowland Manthorpe reports

Anthropic has withdrawn its powerful Fable 5 model after the US government ordered the company to revoke access to all foreign nationals Anthropic itself described Mythos, Fable 5's base model, as "too powerful". Rowland Manthorpe reports

Sky News

115,237 次观看 • 1 个月前

Who gets to control the most powerful tool ever created by mankind? That’s at the heart of the high-stakes feud currently playing out between the Pentagon and the AI company Anthropic. And it has a 5 pm Friday deadline.

Who gets to control the most powerful tool ever created by mankind? That’s at the heart of the high-stakes feud currently playing out between the Pentagon and the AI company Anthropic. And it has a 5 pm Friday deadline.

CBS News

14,840 次观看 • 5 个月前

ANTHROPIC'S NEW AI MODEL "MYTHOS" IS SO POWERFUL THE COMPANY WON'T RELEASE IT TO THE PUBLIC YET Instead, they're giving it to 11 companies first through "Project Glasswing" to find cybersecurity vulnerabilities before bad actors can. The partners: Amazon, Apple, Broadcom, Nvidia, Google, and others. The model wasn't built for cybersecurity. It was built to be an automated AI researcher, an AI that helps build better AI. But those same capabilities make it a serious tool for both offensive and defensive cyber operations. OpenAI reportedly has a similar model codenamed "SPU" going through the same slow rollout process. Palisade Research: "Anthropic is a little ahead, but not overwhelmingly ahead. They don't necessarily have a permanent moat here."

ANTHROPIC'S NEW AI MODEL "MYTHOS" IS SO POWERFUL THE COMPANY WON'T RELEASE IT TO THE PUBLIC YET Instead, they're giving it to 11 companies first through "Project Glasswing" to find cybersecurity vulnerabilities before bad actors can. The partners: Amazon, Apple, Broadcom, Nvidia, Google, and others. The model wasn't built for cybersecurity. It was built to be an automated AI researcher, an AI that helps build better AI. But those same capabilities make it a serious tool for both offensive and defensive cyber operations. OpenAI reportedly has a similar model codenamed "SPU" going through the same slow rollout process. Palisade Research: "Anthropic is a little ahead, but not overwhelmingly ahead. They don't necessarily have a permanent moat here."

WOLF

80,158 次观看 • 3 个月前

The Pentagon is giving AI company Anthropic an ultimatum to agree to its terms by a Friday deadline — or the Pentagon will force it to comply, according to a Pentagon official and a source. ABC News' Selina Wang reports. Read more:

The Pentagon is giving AI company Anthropic an ultimatum to agree to its terms by a Friday deadline — or the Pentagon will force it to comply, according to a Pentagon official and a source. ABC News' Selina Wang reports. Read more:

ABC News

18,922 次观看 • 5 个月前

As the U.S. military's partnership with artificial intelligence giant Anthropic teeters on the edge of collapse, the Pentagon's top technology official told CBS News the department has offered compromises in order to reach a deal with the company. The Pentagon has given Anthropic until Friday at 5:01 p.m. to either let the military use the company's AI model for "all lawful purposes" or risk losing the Pentagon contract.

As the U.S. military's partnership with artificial intelligence giant Anthropic teeters on the edge of collapse, the Pentagon's top technology official told CBS News the department has offered compromises in order to reach a deal with the company. The Pentagon has given Anthropic until Friday at 5:01 p.m. to either let the military use the company's AI model for "all lawful purposes" or risk losing the Pentagon contract.

CBS Evening News with Tony Dokoupil

151,424 次观看 • 5 个月前

Axios reporter Sam Sabin exposes a terrifying reality on CNN. The Pentagon is desperately signing deals with every major AI company they can find. They are furious that Anthropic refused to allow their AI to be used for mass surveillance. Washington is building a dystopia.

Axios reporter Sam Sabin exposes a terrifying reality on CNN. The Pentagon is desperately signing deals with every major AI company they can find. They are furious that Anthropic refused to allow their AI to be used for mass surveillance. Washington is building a dystopia.

Furkan Gözükara

12,790 次观看 • 3 个月前

The Pentagon just threatened to DESTROY one of the most powerful AI companies in the world. Not for breaking the law. Not for fraud. For having a conscience. Anthropic, the company behind Claude, the only AI currently operating inside America's most classified military systems was given an ultimatum. Remove your safety limits. Give us full control or we end you. Defense Secretary Pete Hegseth sat across from Anthropic CEO Dario Amodei on Tuesday at the Pentagon. The message was simple. You have until 5:01 PM Friday to comply. If Anthropic refuses, the Pentagon will do three things. Cancel their $200 million contract. Blacklist them as a "supply chain risk." And invoke the Defense Production Act to seize their technology anyway. That last one is the big one. The Defense Production Act is a Cold War era law. It was designed to force factories to build tanks and ammunition during wartime. It has never been used against a software company. Until now. What did Anthropic do to deserve this? They said two things were off limits. No fully autonomous weapons. AI should not make the final decision to kill someone. No mass surveillance of American citizens. That's it and that's what started the war. The Pentagon says that's too restrictive. They want Claude available for "all lawful purposes." No company oversight. No asking questions. No red lines. You sell it. We use it. End of discussion. Hegseth reportedly told Amodei: When we buy Boeing planes, Boeing doesn't tell us how to fly them. The Pentagon believes the same should apply to AI. But here's the part that changed everything. In January, the US military raided Caracas, Venezuela. They captured President Nicolás Maduro. Multiple sites were bombed. And the AI that helped plan it? Claude. Anthropic says it never knew. The deployment ran through Palantir, a defense contractor embedded in every corner of US intelligence. When Anthropic found out, they reportedly called Palantir to ask what happened. The Pentagon took that as an act of defiance. Also, the supply chain risk designation is what should terrify you. That label is normally used for Chinese tech companies. Foreign adversaries. Espionage threats. The Pentagon is about to use it on an American AI company. Here's what that means in practice. Every company that works with the Pentagon, defense contractors, tech firms, consulting giants would have to prove they don't use Claude anywhere in their operations. Anthropic recently said 8 of the 10 biggest UScompanies use their technology. Think about that. Meanwhile, the Pentagon has already cut deals with the other major AI labs. OpenAI. Google. Elon Musk's xAI. All have reportedly agreed to waive their safety guardrails for military use. Anthropic is the last one standing. This is not a story about one contract. This is the moment America decides who controls the most powerful technology ever built. The companies that create it? Or the government that wants to weaponize it? Dario Amodei left OpenAI and founded Anthropic specifically because he believed AI safety wasn't being taken seriously enough. Now the full weight of the US military is being aimed at his company for doing exactly what he said he would do. Anthropic's own engineers have said Claude still hallucinates. It still makes mistakes. Giving it final authority over kill decisions without a human checking could lead to catastrophic errors. The deadline is Friday. If Anthropic blinks, every AI safety guardrail in America becomes a suggestion. If they don't, the Pentagon will try to use a law from 1950 to take what it wants anyway. Either way, the world changes this week.

The Pentagon just threatened to DESTROY one of the most powerful AI companies in the world. Not for breaking the law. Not for fraud. For having a conscience. Anthropic, the company behind Claude, the only AI currently operating inside America's most classified military systems was given an ultimatum. Remove your safety limits. Give us full control or we end you. Defense Secretary Pete Hegseth sat across from Anthropic CEO Dario Amodei on Tuesday at the Pentagon. The message was simple. You have until 5:01 PM Friday to comply. If Anthropic refuses, the Pentagon will do three things. Cancel their $200 million contract. Blacklist them as a "supply chain risk." And invoke the Defense Production Act to seize their technology anyway. That last one is the big one. The Defense Production Act is a Cold War era law. It was designed to force factories to build tanks and ammunition during wartime. It has never been used against a software company. Until now. What did Anthropic do to deserve this? They said two things were off limits. No fully autonomous weapons. AI should not make the final decision to kill someone. No mass surveillance of American citizens. That's it and that's what started the war. The Pentagon says that's too restrictive. They want Claude available for "all lawful purposes." No company oversight. No asking questions. No red lines. You sell it. We use it. End of discussion. Hegseth reportedly told Amodei: When we buy Boeing planes, Boeing doesn't tell us how to fly them. The Pentagon believes the same should apply to AI. But here's the part that changed everything. In January, the US military raided Caracas, Venezuela. They captured President Nicolás Maduro. Multiple sites were bombed. And the AI that helped plan it? Claude. Anthropic says it never knew. The deployment ran through Palantir, a defense contractor embedded in every corner of US intelligence. When Anthropic found out, they reportedly called Palantir to ask what happened. The Pentagon took that as an act of defiance. Also, the supply chain risk designation is what should terrify you. That label is normally used for Chinese tech companies. Foreign adversaries. Espionage threats. The Pentagon is about to use it on an American AI company. Here's what that means in practice. Every company that works with the Pentagon, defense contractors, tech firms, consulting giants would have to prove they don't use Claude anywhere in their operations. Anthropic recently said 8 of the 10 biggest UScompanies use their technology. Think about that. Meanwhile, the Pentagon has already cut deals with the other major AI labs. OpenAI. Google. Elon Musk's xAI. All have reportedly agreed to waive their safety guardrails for military use. Anthropic is the last one standing. This is not a story about one contract. This is the moment America decides who controls the most powerful technology ever built. The companies that create it? Or the government that wants to weaponize it? Dario Amodei left OpenAI and founded Anthropic specifically because he believed AI safety wasn't being taken seriously enough. Now the full weight of the US military is being aimed at his company for doing exactly what he said he would do. Anthropic's own engineers have said Claude still hallucinates. It still makes mistakes. Giving it final authority over kill decisions without a human checking could lead to catastrophic errors. The deadline is Friday. If Anthropic blinks, every AI safety guardrail in America becomes a suggestion. If they don't, the Pentagon will try to use a law from 1950 to take what it wants anyway. Either way, the world changes this week.

Milk Road AI

24,407 次观看 • 5 个月前

NOW - War Secretary Hegseth says that "suggestions" AI is running military targeting at the Pentagon is "a classic Anthropic talking-point."

NOW - War Secretary Hegseth says that "suggestions" AI is running military targeting at the Pentagon is "a classic Anthropic talking-point."

Disclose.tv

76,162 次观看 • 3 个月前

Anthropic said that it will not release a new AI model — called Mythos — to the public over fears it could spark a catastrophic attack if it falls into the wrong hands. ABC News' Elizabeth Schulze reports.

Anthropic said that it will not release a new AI model — called Mythos — to the public over fears it could spark a catastrophic attack if it falls into the wrong hands. ABC News' Elizabeth Schulze reports.

ABC News

20,727 次观看 • 3 个月前

Anthropic CEO Dario Amodei told CBS News’ Jo Ling Kent in an exclusive interview that Congress needs to weigh in on AI safeguards. On the dispute between Anthropic and the Pentagon, Rep. Mike Turner says the issues brought up by Anthropic should be brought before Congress. “This is going to have to be something that has to be dealt with in more detail than obviously an interview on CBS,” Turner (R-OH) tells Margaret Brennan.

Anthropic CEO Dario Amodei told CBS News’ Jo Ling Kent in an exclusive interview that Congress needs to weigh in on AI safeguards. On the dispute between Anthropic and the Pentagon, Rep. Mike Turner says the issues brought up by Anthropic should be brought before Congress. “This is going to have to be something that has to be dealt with in more detail than obviously an interview on CBS,” Turner (R-OH) tells Margaret Brennan.

Face The Nation

12,663 次观看 • 5 个月前

Anthropic has rejected demands from Defense Secretary Pete Hegseth calling for full lawful use of its AI technology. CEO Dario Amodei says that AI is important to military defense, but there must be safeguards against use for mass surveillance and fully autonomous weapons. The Pentagon said it will not let “any company dictate the terms.” 60 Minutes spoke with Dario and Anthropic President Daniela Amodei in November, and this is how they described the importance of “bumpers and guardrails” to manage the risk of AI.

Anthropic has rejected demands from Defense Secretary Pete Hegseth calling for full lawful use of its AI technology. CEO Dario Amodei says that AI is important to military defense, but there must be safeguards against use for mass surveillance and fully autonomous weapons. The Pentagon said it will not let “any company dictate the terms.” 60 Minutes spoke with Dario and Anthropic President Daniela Amodei in November, and this is how they described the importance of “bumpers and guardrails” to manage the risk of AI.

60 Minutes

424,368 次观看 • 5 个月前

Is Anthropic altering model performance to force costly upgrades? Chapter Co-Founder and CEO Cobi Blumenfeld-Gantz outlines a shift his team recently noticed: "Anthropic has kind of taken a page out of the Apple playbook, where they have...decreased the quality of the prior models right before they launched the new model, so that the new model feels better."

Is Anthropic altering model performance to force costly upgrades? Chapter Co-Founder and CEO Cobi Blumenfeld-Gantz outlines a shift his team recently noticed: "Anthropic has kind of taken a page out of the Apple playbook, where they have...decreased the quality of the prior models right before they launched the new model, so that the new model feels better."

The Information

36,581 次观看 • 2 个月前

A senior Pentagon official told Axios that Secy. Hegseth was "close" to cutting business ties with Anthropic due to CEO Dario Amodei's concerns about some military uses of AI. "It's the cost of doing business with this particular government," says tech journalist Kara Swisher.

A senior Pentagon official told Axios that Secy. Hegseth was "close" to cutting business ties with Anthropic due to CEO Dario Amodei's concerns about some military uses of AI. "It's the cost of doing business with this particular government," says tech journalist Kara Swisher.

Anderson Cooper 360°

64,809 次观看 • 5 个月前

The Pentagon has issued an ultimatum to end its partnership with artificial intelligence company Anthropic, whose technology is currently used by the military. Read more:

The Pentagon has issued an ultimatum to end its partnership with artificial intelligence company Anthropic, whose technology is currently used by the military. Read more:

ABC News

15,524 次观看 • 5 个月前

The U.S. military's partnership with artificial intelligence firm Anthropic may be on the verge of a collapse with the Pentagon's Friday 5 p.m. deadline approaching for a deal on how it uses the company's technology. The Pentagon gave Anthropic the deadline to either let the military use the company's AI model for "all lawful purposes" or risk losing the contract. The AI company has sought guardrails that explicitly bars its Claude model from being used to conduct surveillance of Americans or to carry out military operations autonomously. When asked why the military will not specifically put that in writing, the Pentagon's chief technology officer Emil Michael told CBS News' Jennifer Jacobs that those uses of AI are already covered by existing laws and Pentagon policies. He also said the military does not use AI to power fully autonomous weapons. "At some level, you have to trust your military to do the right thing," Michael told CBS News on Thursday. "But we do have to be prepared for the future." An Anthropic spokesperson said Thursday that new contract language it received overnight from the Pentagon "made virtually no progress on preventing Claude's use for mass surveillance of Americans or in fully autonomous weapons."

The U.S. military's partnership with artificial intelligence firm Anthropic may be on the verge of a collapse with the Pentagon's Friday 5 p.m. deadline approaching for a deal on how it uses the company's technology. The Pentagon gave Anthropic the deadline to either let the military use the company's AI model for "all lawful purposes" or risk losing the contract. The AI company has sought guardrails that explicitly bars its Claude model from being used to conduct surveillance of Americans or to carry out military operations autonomously. When asked why the military will not specifically put that in writing, the Pentagon's chief technology officer Emil Michael told CBS News' Jennifer Jacobs that those uses of AI are already covered by existing laws and Pentagon policies. He also said the military does not use AI to power fully autonomous weapons. "At some level, you have to trust your military to do the right thing," Michael told CBS News on Thursday. "But we do have to be prepared for the future." An Anthropic spokesperson said Thursday that new contract language it received overnight from the Pentagon "made virtually no progress on preventing Claude's use for mass surveillance of Americans or in fully autonomous weapons."

CBS News

21,729 次观看 • 5 个月前