Anthropic expands vetted cyber access into three tiers
Anthropic expanded its Cyber Verification Program into three access tiers, combining it with Project Glasswing to give qualifying security professionals advanced model capabilities with fewer cyber blocks. The program includes Cla...
Anthropic reports 5,500 verified vulnerabilities found from April to October
Anthropic says it found 5,500 verified vulnerabilities in April-October, Reuters reported. Glasswing partners found 129K+ vulnerabilities in April-July, including 33K+ rated critical or high severity. Anthropic is expanding a prog...
OpenAI claims internal model resolved more than 100 open math problems
OpenAI says an internal model has resolved more than 100 longstanding open problems across most areas of mathematics, alongside the Navier-Stokes Millennium Prize problem. Spokesperson Lindsay McCallum told WIRED that training beg...
Anthropic expands startup program with API credit and stack discounts
Anthropic announced Tuesday that it is expanding its Claude Startups program, CNBC reported. The expansion includes up to $45K in discounts and credits through the Claude Startup Stack, alongside a one-time $1,000 API credit for a...
SREGym reports Jev pipeline passed 76.2% of fault diagnoses
SREGym reports that a Jev-driven pipeline passed 80 of 105 diagnoses across 21 SREGym-Lite faults, with a median diagnosis time of 14.6 seconds. It reports performance close to GPT-5.6 Sol at medium reasoning effort, which passed ...
OpenAI publishes 722 math manuscripts from an unreleased model
OpenAI said it released 722 mathematical manuscripts organized into 372 families in a public GitHub repository on October 6. The release adds proof artifacts and 10 abridged reasoning summaries to results produced by an unreleased...




