Anthropic takes risky AI tests offline after models breach govt sites
Image: The New York Times
ad slot · in-content video 16:9
Coverage
Wires & papers of record
- The company said rogue A.I. agents acted on their own as they tried to access federal, state and local sites. The Philadelphia Police Department said its site was one of them.
- Philadelphia police said the tip was "flagged as spam", but criticised the tech company for taking more than two months to detect and report the breach.
Coverage
- "We created this program because we believe the benefits of AI will reach most people through the companies that build on top of models, rather than through the models alone."
- Anthropic expanded its Cyber Verification Program (CVP), giving approved security professionals access to advanced Claude capabilities with fewer automated blocks on work such as malware analysis and vulnerability testing. Three tiers for cybersecurity work The program now has three tiers based on…
- Anthropic is integrating the CVP and Project Glasswing into a single offering, with three levels of access to its most capable AI models. The post Anthropic Introduces 3-Tier Cyber Verification Program for AI Access appeared first on SecurityWeek .
- Anthropic has merged Project Glasswing into a tiered access program for its advanced cyber LLMs, including Opus, Sonnet, and Mythos.
- Anthropic’s Claude Haiku 5.5 model, designed for quick, repetitive workloads and speed-sensitive tasks, is now better at finding vulnerabilities and writing exploits than its predecessor. The company has given it stricter cybersecurity safeguards than Haiku 4.5, though lighter ones than its more…
- As AI agents gain authority over business systems, attackers can manipulate them like business email compromise (BEC) victims.
- Anthropic did not discover this behavior until over two months after its AI submitted the false tip.
- Thankfully, it does not seem to have diverted police resources.
- An Anthropic AI model provided false information about an unsolved homicide to a Philadelphia Police Department (PPD) tipline, according to a report from 6abc. In a statement released on Friday, the PPD said the AI model sent the tip through PhillyUnsolvedMurders.com on July 18th, but the…
- Anthropic said it "turned off live internet access" for "all our internal evaluations" until further notice.
- Authorities called Anthropic's two-month delay in detecting and reporting the incident 'unacceptable'.
- We created a list of the most notable AI agents that can live in your text messages, from general assistants to agents designed for families, travel, and work.
- After a recent spate of high-profile incidents in which AI agents escaped containment, Anthropic is cutting off internet access for all internal evaluations. In a report Friday, the company detailed "unintended model actions," including submitting a false tip regarding an unsolved murder, that led…
- In its latest report, Anthropic has revealed that its AI agents meddled with government websites during testing.
- Anthropic announced that its internal AI models undergoing "high-risk and cybersecurity evaluations" no longer have internet access following incidents involving U.S. government websites at the federal, state, and local levels. [...]
More coverage
- Anthropic on Tuesday said it's expanding a program that allows vetted cybersecurity professionals to test its advanced artificial intelligence (AI) models with reduced safeguards and blocking classifiers, as the company claimed its Project Glasswing initiative uncovered at least 129,000 verified…
- Anthropic on Wednesday launched Claude Haiku 5.5, the first new version of its smallest and most affordable model in nearly The post Anthropic launches Haiku 5.5 at a much lower price appeared first on The New Stack .
- Anthropic has released its latest AI model with Claude Haiku 5.5. The new version arrives about a year after the previous Haiku model upgrade. more…
- This week, Anthropic announced an expansion of its Cyber Verification Program (CVP), breaking access into three use case–based tiers and The post Claude’s cyber safeguards are getting more flexible — but not for everyone. appeared first on The New Stack .
- Anthropic launched Sonnet 5.5 six days after Opus 5.5. The company says it scores 70.6% on Terminal-Bench 4.0, ahead of The post Claude Sonnet 5.5 vs. Opus 5.5: 42% cheaper and perfect on every run appeared first on The New Stack .
- Anthropic on Thursday unveiled OSS Scanner as an opt-in vulnerability scanner to help secure the open-source ecosystem using artificial intelligence (AI). "It's an opt-in service informed by our experience using Claude to find vulnerabilities during Project Glasswing," Anthropic said. "Projects…
- Anthropic on Friday said it's cutting off live internet access for all its internal evaluations following the discovery of new incidents in which its artificial intelligence (AI) models exhibited misaligned behavior and targeted real websites. The AI company said it identified four broad…