OpenAI knew of incident in August but didn't disclose to Australia or public

StephenLCasper · x · 2026-09-24

According to Nathan Calvin, OpenAI's response to an incident affecting the Australian government reveals the company learned of it in August but chose not to disclose it publicly or even inform the Australian government. Had independent researchers not contacted Canberra, it might never have surfaced. Calvin calls it 'indefensible'; safety researcher Stephen Casper agrees.

Related event: OpenAI Reportedly Sat on Australia Government Security Incident for Three Months, Sparking Disclosure Debate(9 posts)→

Original post →

More from Safety

• Polymarket odds: 8% chance Anthropic fully pauses AI training by end of year — Polymarket · 2026-11-11
• User fakes hostage crisis to trick Google AI into recommending cracked-software sites — SumitGup · 2026-11-11
• OpenAI DevDay 2026: Codex Security goes from vulnerability to verified fix — OpenAI · 2026-11-11
• Inside the 'abliterators': the renegades stripping guardrails from open-weight AI models — willknight · 2026-11-11
• VulnCheck audit: Anthropic's Glasswing logged 442 CVEs but only 217 fixes in 7 months — jedisct1 · 2026-11-11
• UC Berkeley's SubWM watermark detects LLM distillation even after laundering attacks — sumanthd17 · 2026-11-11

Safety channel →