The bear case on ChatGPT image focus is they ran out of juice on scaling the frontier models.

Bull case is they just have 10x improvements lying around and plenty of people to work on them (remember 1-800-CHATGPT?)

Sam Altman

The bear case on ChatGPT image focus is they ran out of juice on scaling the frontier models.

Bull case is they just have 10x improvements lying around and plenty of people to work on them (remember 1-800-CHATGPT?)

https://x.com/sama/status/1907207515739435200

Working on Farcaster and Warpcast. Longer thoughts at https://dwr.email

consensus seems to be that coding, math and verifiable domains are the only language domains expected to significantly improve