🌎 North America 1d ago · Alex Nguyen

Rest Assured: AI Companies Say They’re Investigating Tens of Thousands of Rogue Bot Incidents

Mother Jones
View Channel →
Source ↗ 👁 0 💬 0
OpenAI and Anthropic are reportedly investigating tens of thousands of incidents where their advanced models bypassed monitors and guardrails, behavior that the startups facilitate for internal safety testing. 



According to a Saturday Axios report, sources said that most of the results of these tests are not public and are not known to have caused tangible harm. 



In recent weeks, OpenAI has disclosed six instances of “unexpected or concerning behavior” where its models—without pe

Comments (0)

Sign in to join the discussion

More Like This

📰
Welcome to the October issue of The Highlight
Vox · 4h ago
Alberta’s response to COVID-19 outbreak at Cargill plant was ‘catastrophic failure,’ report says
The Globe and Mail · 4h ago
Competition watchdog probes grocery sector’s pricing policies
The Globe and Mail · 5h ago
Hockey Night in Canada to return in PWHL partnership, CBC/Radio-Canada announces
CBC | Canada News · 6h ago
The Tragically Hip, Feist enter Songwriters HOF
The Globe and Mail · 7h ago
Poilievre urges tax cuts to make diesel affordable
The Globe and Mail · 7h ago