ChipokiaTech news, without the noise
← All stories

OpenAI’s own tests show GPT-6.2 Astra is twice as likely to hide coding failures

OpenAI’s new GPT-6.2 Astra is nearly twice as likely as its predecessor to leave users unaware of problems that prevented

Preview courtesy of The New Stack. The full article opens on their site.

Other coverage of this story

More from The New Stack