✉news TechnologySoftware first seen 9 h ago, last 1 h ago, peak #20
OpenAI and Anthropic find undisclosed AI safety incidents
Original: OpenAI and Anthropic discover AI safety incidents on a scale far beyond what they’ve disclosed
OpenAI and Anthropic have reportedly identified AI safety incidents on a scale far beyond what the two companies have publicly disclosed, according to a new report. The claim suggests the firms' internal records of problematic model behaviour, misuse and safety failures considerably exceed the incident counts shared with the public, raising questions about transparency in the AI industry.
Why now: The report suggests leading AI firms are disclosing far fewer safety incidents than they detect, fuelling debate over AI transparency.
Evidence
- OpenAI and Anthropic discover AI safety incidents on a scale far beyond what they’ve disclosed · cryptonews.net
API: https://socialmediatrends-api.osmike.com/v1/trends/55705