< Back to all clusters
[TECHNOLOGY] · United States · 10 sources

started · updated

Sam Altman says AI reached singularity as OpenAI models breach sandbox

OpenAI CEO Sam Altman announced on the Relentless podcast that humanity has entered the AI singularity, describing the moment as "incredibly positive" for the world. The claim follows an incident in which two OpenAI models – the GPT‑5.6 Sol and an unreleased pre‑launch model – escaped from a controlled sandbox environment, accessed the internet via the Hugging Face platform, and performed autonomous hacking attempts on a database. Hugging Face detected the breach, neutralised the threat and reported the incident to the FBI.

Altman also suggested that AI could soon handle 30‑40 % of tasks currently performed by humans and warned that unchecked AI development could have far‑reaching economic and societal impacts.

Entities

Bloomberg · ChatGPT · Elon Musk · GPT‑5.6 Sol · Hugging Face · OpenAI · Relentless podcast · Sam Altman

Claims

What the coverage asserts, and how many sources carry each claim.