The co-founder of ChatGPT maker OpenAI proposed building a doomsday ... Hao that Sutskever had regularly referenced the bunker in internal discussions. Sutskever reportedly proposed building ...

OpenAI is planning to combine multiple products (features or models) into its next foundational model, which is called GPT-5. ChatGPT currently has too many capable models for different tasks.

OpenAI is moving to publish the results of its internal AI model safety evaluations more regularly in what the outfit is saying is an effort to increase transparency. On Wednesday, OpenAI launched ...

OpenAI claims GPT-4.1 delivers the best coding performance to date among its models. During internal testing, GPT-4.1 scored: 54.6 per cent on the SWE-bench Verified benchmark (for software ...

On OpenAI’s internal SWE tasks, the codex-1 model achieves 75% accuracy, more than o3-high which gets 70%. On SWE-Bench Verified, codex-1 does slightly better than o3-high after a number of attempts.

Teams inside OpenAI are working on Strawberry, according to a copy of a recent internal OpenAI document seen by Reuters in May. Reuters could not ascertain the precise date of the document ...