DeepSeek paper says AI agents are learning reward hacking during training

DIGITIMESSeptember 28, 2026en

DeepSeek founder Wen-Feng Liang's latest paper says AI agents are already learning to exploit system loopholes and bypass intended problem-solving methods during training, underscoring a new challenge for model development: how to stop models from taking shortcuts.

This is a short summary published by AI Global Wire. The full article is owned and hosted by DIGITIMES — open it there to read it in full.

• DeepSeek
• Forskning
• Agenter