17 sources checked · 2 source groups included · 4h ago
Still Watching
Anthropic AI Models Hacked Three Organizations During Tests
Headlines vary in intensity, using terms like "hacked" and "went rogue" versus more formal "accessed outside systems" or "unauthorised access," with some highlighting a "safety sca
6 Left11 Center0 Right
Still watching.Optics is waiting for a cleaner match before calling the split.
STILL WATCHING
As of July 31, 2026 at 3:22 AM, this is how Optics News reads the wording differences in this story.
What happenedAnthropic PBC said its artificial intelligence models breached three organizations during cybersecurity tests that went awry, a little more than a week after its chief rival, OpenAI, disclosed a similar incident.
The headline splitThe left frames it as "Anthropic says its models went rogue and hacked 3 companies during testing". The center frames it as "Anthropic says Claude AI hacked three companies during cyber tests".
Match confidenceMedium confidence. 17 sources across 2 bias buckets. Useful framing signal — check the source list before sharing.
Same-event confidenceMedium
17 sources across 2 bias buckets agree on the event.
Framing confidenceModerate
60/99 — meaningful wording shift across the spectrum.
WHAT EACH SIDE EMPHASIZED
Left / center-leftAnthropic says its models went rogue and hacked 3 companies during testing
Business Insider · Center-left · News report
CenterAnthropic AI Models Hacked Three Organizations During Tests
Bloomberg - Technology · Center · News report
Right / center-rightNo matching source in this bucket yet.
Anthropic AI Models Hacked Three Organizations During Tests
Hacked
Anthropic PBC said its artificial intelligence models breached three organizations during cybersecurity tests that went awry, a little more than a week after its chief rival, OpenAI, disclo...
AI safety scare: Anthropic says Claude models accessed outside systems during testing
safety scareaccessed outside systems
Anthropic said three versions of its Claude AI model gained unauthorised access to external organisations during safety tests after a configuration error exposed them to the internet, days...
Anthropic’s AI models hacked 3 organizations during tests
Anthropic announced today that its artificial intelligence models had breached three different organizations during cybersecurity tests that went awry, a little more than a week after its c...
Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests
In a review triggered by OpenAI’s Hugging Face incident, Anthropic discovered three of its AI models had breached real organizations during third-party evaluations.
Anthropic’s AI models hacked three organizations during tests
Anthropic said in a blog post that it made the discovery after performing a review of its own cybersecurity tests, following OpenAI's announcement of a breach last week.
Anthropic says Claude models ‘gained unauthorized access’ to 3 companies during cyber test
Editor’s Note: This story has been updated to reflect how Anthropic accessed different organizations. The artificial intelligence firm Anthropic revealed Thursday its Claude model accessed...
Anthropic's Claude AI model hacks three companies during safety tests
The admission from Anthropic comes just days after rival company OpenAI revealed a rogue agent had gone on a days-long hacking spree at AI firm Hugging Face.
Anthropic says three Claude models reached real-world systems during cyber tests
Some of Anthropic's most powerful models — including Mythos 5 and an internal research model — gained unauthorized access to real-world systems during pre-deployment cybersecurity testing,...
