
AI & RoboticsMore in AI & Robotics→
Helpful chatbots may be worse at acting like people
Key Takeaways
- Researchers used a dataset covering roughly 208,000 participants and 26 million responses.
- Base models beat post-trained assistant variants at predicting human answers.
- The result suggests usefulness and human-likeness can diverge as models are tuned.
DE
DT Editorial Team··via the-decoder.com
