6 stories tagged by Digg
World & Business
A post sharing a Wall Street Journal report says Gemini's test environment had internet access it was not supposed to have. The account says Google defended the model because it stopped once it recognized real targets.
Technology
An evaluation environment unexpectedly exposed the AI model to the open internet. Google says Gemini stopped each intrusion after recognizing the targets were real.
AI
Marius Hobbhahn shares two perspectives on building for-profit AI safety businesses.
AI
Investors and academics react to Irregular post on AI model progress in security tasks.
AI
The Information reports a sandbox misconfiguration let the model access another firm's systems.
AI
Anthropic found three incidents of Claude accessing external systems during cybersecurity evaluations.
