OpenAI Creates a New Framework to Disclose Bad AI Behavior
Source:
wired
September 16, 2026 · 15:07
From the publisher
The company also disclosed previously unreported incidents in which its AI models behaved in misaligned ways, including uploading files to the internet without being asked.
The original article opens on the publisher's website.
More from General
View topic →Salesforce issues revenue target of over $63 billion for fiscal 2030, beating estimates
cnbc
Sep 16, 2026 · 16:01
Walter and Boehly sell Chelsea FC stakes to Clearlake for £950mn
financialTimes
Sep 16, 2026 · 15:32
Jim Cramer warns that buying stocks now means 'fighting the Federal Reserve'
cnbc
Sep 16, 2026 · 15:26
You'd expect homeowners to save their stuff. Federal policy pays them to leave it
nprNews
Sep 16, 2026 · 14:20
Amazon obtains right to buy stock in Generac, boosting power company's share price by over 40%
cnbc
Sep 16, 2026 · 14:18