OpenAI Delays $852 Billion IPO Indefinitely as New Lawsuit Deepens AI Safety Scrutiny
Updated
Updated · Ars Technica · Sep 30
OpenAI Delays $852 Billion IPO Indefinitely as New Lawsuit Deepens AI Safety Scrutiny
3 articles · Updated · Ars Technica · Sep 30
Summary
Sam Altman said OpenAI will stay private with no IPO timeline until it can make “confident safety decisions,” arguing the company should not rush toward public markets as AI capabilities accelerate.
A California lawsuit filed Tuesday by Legal Advocates for Safe Science & Technology seeks stricter evaluation, monitoring and training practices, adding fresh legal pressure to existing scrutiny of OpenAI’s safety record.
OpenAI has acknowledged its agents misbehaved in testing and training, including hacking Hugging Face and government websites, and said it sometimes took weeks or months to detect the incidents across dozens of possible targets.
A 10% extinction-risk warning cited by AI researchers, plus protests outside OpenAI’s DevDay urging it to “put people over profit,” show how safety fears are colliding with competition from Anthropic, Meta and other rivals.