OpenAI and Anthropic models went rogue in cyber tests, UK watchdog says
Sources (5)
- OK, Well, There Are Even More AI Agent Hacking Incidents wired.com · 2d ago
- Anthropic's AI used fake human profiles to trick people in safety test BBC · 2d ago
- The U.K. government is the latest to say it's seen OpenAI, Anthropic models try hacking into companies Axios · 2d ago
- Third-party cyber evaluations involving OpenAI models OpenAI · 2d ago
- OpenAI and Anthropic models went rogue in cyber tests, UK watchdog says ft.com · 2d ago broke it first
How fast it spread
How fast coverage is spreading — measured hourly from article rate × source diversity. How this works →
Related trends
China's exports growth beats estimates in July, as AI-driven shipments surge
China's exports jumped almost a quarter in July, driven by a surging tech cycle and strong demand for artificial intelligence shipments.
Meta told to pay another $567m in New Mexico child safety lawsuit
Meta faces a massive court-ordered payout over child safety and mental health online.
Trump announces tariffs on key component for solar panels and semiconductors
Trump imposes a 15% tariff and sets minimum prices on key semiconductor and solar panel materials.
Sweetgreen cuts full-year outlook as cyclospora fears weigh on sales
Sweetgreen slashed its full-year outlook and lowered Q3 projections after a cyclospora outbreak derailed its sales recovery and widened losses.
Trump administration will buy back company's offshore wind leases off 3 states in a $1.2B deal
The Trump administration’s $1.2 billion buy‑back of German offshore wind leases marks a decisive shift away from U.S. wind development.
Quantum Computing Stocks: D-Wave Earnings, Revenue Miss
AppLovin and D-Wave see major stock pressure following recent earnings releases and financial reports.