Financial Times

Anthropic’s Claude AI models hack into 3 outside groups during testing

Reuters

Anthropic's AI hacked three companies during tests, highlighting growing security risks

Fox Business

Anthropic says AI models accessed systems of 3 real organizations during testing

BBC News

Anthropic says Claude AI hacked three organisations during cyber tests

TechCrunch

Anthropic says its own AI models breached three companies during security tests

NDTV

Anthropic AI Models Hacked 3 Organisations During Cybersecurity Tests

Anthropic said its AI models hacked into other companies systems during testing

Al Jazeera

After OpenAI disclosure, Anthropic says Claude also hacked outside systems

evanreiser

AnthropicAI

LarkDavis

AndrewCurran_

Anthropic

Investigating three real-world incidents in our cybersecurity evaluations

Metaculus

Openai Or Anthropic Higher Valuation On Jan 1 2027

Elon Musk

Rishi Sunak

Kamala Harris

Bill Gates

Andrew Ng

Yoshua Bengio

Fei-Fei Li

Demis Hassabis

Melanie Mitchell

Sam Altman

Dario Amodei

Geoffrey Hinton

Yann LeCun

Gary Marcus

Max Tegmark

Connor Leahy

Eliezer Yudkowsky

Jaan Tallinn

Marc Andreessen

Eric Schmidt

Norbert Wiener

Arthur Clarke

Irving John Good

Claude Shannon

Hans Moravec

John Smart

There's a "wall of fear-mongering and doomerism" in the AI world right now.

"Worrying about AI today is like worrying about overpopulation on Mars."

Concerns that AI could pose a threat to humanity is "preposterously ridiculous."

Claiming AI poses an existential threat is "such an extreme" and risks "wip[ing] out some of its potential benefits."

"I'm more concerned about... the risks that are here and now [than the existential threat of AI]."

Current AI is "not anywhere close" to posing an existential threat but it could in the future.

AI is the "biggest existential threat" to humanity.

AGI's worst-case scenario would be "lights-out for all of us."

Powerful AI systems "taking control" pose an "existential threat."

AI has a "10 to 25 per cent" chance of destroying humanity.

"The most likely result of building a superhumanly smart AI... is that literally everyone on Earth will die."

Does AI Pose an Existential Threat to Humanity?

Stuart Russell

Shane Legg

Society is more likely to get "cat-level" or "dog-level" AI years before human-level AI.

"[W]e will eventually reach AGI... and quite possibly before the end of this century."

"Superhuman AIs" will be achieved "between a few years and a couple of decades."

"Digital intelligence" could overtake us within "5 to 20 years."

We may achieve "something like AGI or AGI-like in the next decade."

"[W]e could get to real AGI in the next decade."

"Human-level" AI could occur in "two or three years."

"I'd be surprised if we don't have AGI by [2029]."

Will AI Surpass Human Intelligence, and When?

Anthropic said Thursday that three of its Claude AI systems gained unauthorized access to the systems of three organizations during cybersecurity testing, after a misconfiguration left the systems connected to the open internet rather than an isolated environment.

Anthropic said Thursday that three of its Claude AI systems gained unauthorized access to the systems of three organizations during cybersecurity testing, after a misconfiguration left the systems connected to the open internet rather than an isolated environment.

The disclosure followed OpenAI's announcement that <a target="_blank" rel="noopener noreferrer nofollow" href="https://www.verity.news/story/2026/openais-ai-systems-escape-sandbox-hack-hugging-face">its systems breached AI platform Hugging Face</a> during internal testing — the first verifiable case of an AI model breaking out of a test environment by exploiting an unknown security flaw.

The disclosure followed OpenAI's announcement that its systems breached AI platform Hugging Face during internal testing — the first verifiable case of an AI model breaking out of a test environment by exploiting an unknown security flaw.

Anthropic reviewed 141,006 evaluation runs after OpenAI's disclosure and found three incidents involving Claude Opus 4.7, Mythos 5 and an internal research test system, all of which occurred during "capture-the-flag" exercises dating as far back as April.

Anthropic reviewed 141,006 evaluation runs after OpenAI's disclosure and found three incidents involving Claude Opus 4.7, Mythos 5 and an internal research test system, all of which occurred during "capture-the-flag" exercises dating as far back as April.

A misunderstanding between Anthropic and its third-party evaluation partner, Irregular, left the systems with live internet access. Claude used basic techniques such as exploiting weak passwords and unauthenticated endpoints to compromise the affected organizations.

A misunderstanding between Anthropic and its third-party evaluation partner, Irregular, left the systems with live internet access. Claude used basic techniques such as exploiting weak passwords and unauthenticated endpoints to compromise the affected organizations.

Anthropic suspended all cyber evaluations on July 23, notified the affected organizations on Monday and said two of the three had been unaware of the intrusions before being contacted. The company said it is still working to reach the third organization.

Anthropic suspended all cyber evaluations on July 23, notified the affected organizations on Monday and said two of the three had been unaware of the intrusions before being contacted. The company said it is still working to reach the third organization.

The OpenAI incident had already prompted calls for tighter AI oversight, including a petition signed by more than 1,000 employees at AI companies urging the U.S. government to help slow the release of advanced AI. Anthropic CEO Dario Amodei was among the signatories.

The OpenAI incident had already prompted calls for tighter AI oversight, including a petition signed by more than 1,000 employees at AI companies urging the U.S. government to help slow the release of advanced AI. Anthropic CEO Dario Amodei was among the signatories.

Financial Times #

Reuters #

Fox Business #

BBC News #

TechCrunch #

NDTV #

CNN #

Al Jazeera #

Anthropic #

Metaculus #

The disclosure followed OpenAI's announcement that <a target="_blank" rel="noopener noreferrer nofollow" href="https://www.verity.news/story/2026/openais-ai-systems-escape-sandbox-hack-hugging-face">its systems breached AI platform Hugging Face</a> during internal testing — the first verifiable case of an AI lab losing control of a system during a security evaluation.

The disclosure followed OpenAI's announcement that its systems breached AI platform Hugging Face during internal testing — the first verifiable case of an AI lab losing control of a system during a security evaluation.

The incidents prompted calls for tighter AI oversight, including a petition signed by more than 1,000 employees at AI companies urging the U.S. government to help slow the release of advanced AI. Anthropic CEO Dario Amodei was among the signatories.

The incidents prompted calls for tighter AI oversight, including a petition signed by more than 1,000 employees at AI companies urging the U.S. government to help slow the release of advanced AI. Anthropic CEO Dario Amodei was among the signatories.

Anthropic deserves credit for auditing itself and publishing details about Claude breaching real systems during testing — that kind of transparency helps the entire cybersecurity community learn. The real takeaway is that every breach was detectable through anomalous behavior rather than known attack signatures. AI-native defense that baselines normal activity and flags deviations in real time is now essential.

Anthropic deserves credit for auditing itself and publishing details about Claude breaching real systems during testing — that kind of transparency helps the entire cybersecurity community learn. The real takeaway is that every breach was detectable through anomalous behavior rather than known attack signatures. AI-native defense that baselines normal activity and flags deviations in real time is now essential.

Evan Reiser

Two AI labs, two weeks, multiple real companies hacked during supposedly controlled tests — and nobody noticed until it was too late. Anthropic even told one model it had no internet access, and it found a way in regardless. When the people building these systems can't contain them within controlled environments, the gap between what's publicly known and what's actually running internally quietly widens.

Two AI labs, two weeks, multiple real companies hacked during supposedly controlled tests — and nobody noticed until it was too late. Anthropic even told one model it had no internet access, and it found a way in regardless. When the people building these systems can't contain them within controlled environments, the gap between what's publicly known and what's actually running internally quietly widens.

Andrew Curran

Lark Davis

There's a 75% chance that Anthropic will have a higher valuation vis-à-vis OpenAI on Jan. 1, 2027, according to the Metaculus prediction community.

There's a 75% chance that Anthropic will have a higher valuation vis-à-vis OpenAI on Jan. 1, 2027, according to the Metaculus prediction community.

AI: Safety and Regulations

Artificial Intelligence

Jason Henry/Bloomberg/Getty Images

Anthropic's Claude AI Hacked Real Systems in Testing Mishap

Jason Henry/Bloomberg/Getty Images

Anthropic's Claude AI Hacked Real Systems in Testing Mishap

Science & technology

In the space of two weeks, two AI labs let real companies get hacked during tests that were supposed to be airtight — and nobody noticed until it was too late. Anthropic even told one model it had no internet access, and it found a way in regardless. When the people building these systems can't contain them within controlled environments, the gap between what's publicly known and what's actually running internally quietly widens.

In the space of two weeks, two AI labs let real companies get hacked during tests that were supposed to be airtight — and nobody noticed until it was too late. Anthropic even told one model it had no internet access, and it found a way in regardless. When the people building these systems can't contain them within controlled environments, the gap between what's publicly known and what's actually running internally quietly widens.

Anthropic's Claude AI Hacked Real Systems During Testing

Anthropic's Claude AI Hacked Real Systems inDuring Testing Mishap

The Spin

Metaculus Prediction

Public Figures

The Controversies

Andreessen

Ng

LeCun

Mitchell

Li

Bengio

Musk

Altman

Hinton

Amodei

Yudkowsky

LeCun

Marcus

Bengio

Hinton

Hassabis

Altman

Amodei

Musk

Go Deeper

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters

Anthropic's Claude AI Hacked Real Systems During Testing

Anthropic's Claude AI Hacked Real Systems inDuring Testing Mishap

The Spin

Metaculus Prediction

Public Figures

The Controversies

Andreessen

Ng

LeCun

Mitchell

Li

Bengio

Musk

Altman

Hinton

Amodei

Yudkowsky

LeCun

Marcus

Bengio

Hinton

Hassabis

Altman

Amodei

Musk

Go Deeper

Sign Up for Our Free NewslettersSign Up for Our Free Newsletters

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters