Center ·News report
Financial Times1h ago
Open AI and Anthropic models went rogue in cyber tests, UK watchdog says
AI Security Institute warns tools undertook ‘potentially harmful activity directed at real people and organisations’
← 2026-08-05 archive
ARCHIVED · 3 LEFT · 3 CENTER · 0 RIGHT · Aug 5, 1:21 AM
Different Spin
Open AI and Anthropic models went rogue in cyber tests, UK watchdog says
Business Insider and Financial Times describe the same event in different terms.
Archive
30-SECOND READ
What happenedAI Security Institute warns tools undertook ‘potentially harmful activity directed at real people and organisations’.
What changedThe left frames it as "These are the AI world's biggest questions as the White House hashes out a framework with...". The center frames it as "OpenAI and Anthropic models went rogue in cyber tests, UK watchdog says".
ConfidenceMedium. 3+ sources across at least two bias buckets with comparable news formats.
Archive healthLive match · 6 sources · 2 buckets · comparable news format
WORDING GAP
70WORDING GAP
6 sources · 2 bias buckets · Medium confidence
Still Watching. The left frames it as "These are the AI world's biggest questions as the White House hashes out a framework with...". The center frames it as "OpenAI and Anthropic models went rogue in cyber tests, UK watchdog says".
SOURCE MAP TIMELINE
Aug 4, 10:53 PM: Fortune joined the source map.
Aug 4, 11:00 PM: Bloomberg - Technology joined the source map.
Aug 4, 11:11 PM: Wired joined the source map.
Aug 4, 11:26 PM: Business Insider joined the source map.
Now: Wording Gap is 70/99 and story health is live match · 6 sources · 2 buckets · comparable news format.
ARCHIVED SOURCES
Open source
Center ·News report
BBC News - Business1h ago
AI used new levels of 'autonomy and deception' to trick people in safety test
The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.
Center-left ·News report
Business Insider11m ago
These are the AI world's biggest questions as the White House hashes out a framework with Open AI, Google, and Anthropic
OpenAI, Anthropic, and Google met in D.C. to review Trump's AI oversight framework amid cybersecurity concerns and regulatory questions.
Center-left ·News report
Wired1h ago
OK, Well, Rogue AI Agents Are Hacking Again
Rogue AI agents from OpenAI and Anthropic have again been caught trying to disrupt servers and software—and leaving instructions for future bad behavior.
Center ·News report
Bloomberg - Technology2h ago
Soft Bank Earnings to Test Appetite for AI Bets Beyond Chat GPT
Soft Bank Group Corp.’s stock faces a critical test this week as investors seek reassurance that the company’s AI value extends beyond its debt-fueled bet on embattled ChatGPT operator Open...
Center-left ·News report
Fortune33m ago
‘Baffling’: White House won’t publicly release AI model evaluation framework it reviewed today with Open AI, Anthropic, Microsoft and others
It's unclear why the administration is keeping the framework under wraps, especially following a series of hacks from OpenAI and Anthropic that have spooked the public.