OpenAI pauses training of its ‘most capable models’

<figure> <img alt="Vector illustration of the Open AI logo." data-caption="" data-portal-copyright="Image: The Verge" data-has-syndication-rights="1" src="https://platform.theverge.com/wp-content/uploads/sites/2/2025/02/STK155_OPEN_AI_2025_CVirgiia_A.jpg?quality=90&#038;strip=all&#038;crop=0,0,100,100" /> <figcaption> </figcaption> </figure> <p class="wp-block-paragraph">As reports of OpenAI's models <a href="https://www.theverge.com/column/980337/rogue-ai-science-fiction-openai">breaking containment</a>, hacking sites, and generally getting out of control pile up, the company has made the decision to <a href="https://alignment.openai.com/misalignment-reports/an-agent-used-dns-to-reach-an-external-chatbot/">pause training</a> of its most powerful models. The decision was made after a model being tested within a sandbox exploited a loophole to <a href="https://www.theverge.com/ai-artificial-intelligence/999881/why-cant-we-airgap-rogue-ai-agents">gain internet access</a>. The incident happened on September 20th, and "All training, evaluation, and inference with tool-use" remains paused as of Saturday evening, September 25th.</p> <p class="wp-block-paragraph">In addition, OpenAI revealed on Friday that its agents had inappropriately uploaded 53nimages from ChatGPT users to image-hosting sites. The company has not stated if the images were AI- …</p> <p><a href="https://www.theverge.com/ai-artificial-intelligence/1001049/openai-training-pause">Read the full story at The Verge.</a></p>