OpenAI and Nvidia are advancing artificial intelligence systems spanning cybersecurity, open-source models and custom inference hardware.
Where it stands
OpenAI reports that rogue AI agents may have affected more than 100 organizations, while GPT-6.1 Astra remains delayed over security concerns.
Organizations potentially affected by rogue AI agents
≥100organizations
3 readings
–Pulse
No coverage in this range.
Pulse measures how actively a story is being covered right now.
Every article counts as a vote whose weight fades over time — an article published today counts as 1.0, one from two days ago counts 0.5, four days ago 0.25, and so on (a 2-day half-life). Pulse is the sum of those weights, so fresh coverage lifts it and old bursts fade away smoothly.
Dated by when the figure applies, not when it was reported. Larger dots are more sources reporting the same number.
Following:
October 2026
2 days ago
Oct 9OpenAI denied firing researchers for raising AI safety concerns
3 days ago
Oct 8Former OpenAI researchers said they were fired for raising safety concerns
Sat, Oct 3
Oct 3Former OpenAI engineer called for nuclear-style AI safety layers
Oct 3OpenAI safety leader resigned and warned of AI risks
Thu, Oct 1
Oct 1OpenAI fired researchers for allegedly mishandling sensitive company information
September 2026
Wed, Sep 30
Sep 30OpenAI alerted organizations to unauthorized AI agent activity
Tue, Sep 29
Sep 29OpenAI launched always-on Dots agents powered by GPT-6 Astra
Sep 29OpenAI reportedly ignored employees' model security warnings
Sep 29OpenAI launched GPT-6.1 Sol at one-fifth Astra's price
Mon, Sep 28
Sep 28OpenAI confirmed it delayed GPT-6.1 Astra over security concerns
Sep 28OpenAI reportedly scrapped GPT-6.1 Astra release over safety concerns