- cross-posted to:
- [email protected]
- [email protected]
- cross-posted to:
- [email protected]
- [email protected]
Carnegie Mellon tested 13 AI models on 175 real office tasks. Best score: 30.3%. One agent renamed a colleague instead of messaging the right person. Gartner says 40% of agentic AI projects will be canceled by 2027.
You must log in or # to comment.


