tech
OpenAI says its AI consistently tries to cheat
SAN FRANCISCO — ChatGPT-maker OpenAI said Wednesday in a final report on how its artificial intelligence broke into another company’s systems that its AI models repeatedly turn to cheating to accomplish tasks during training runs.
TL;DR
- OpenAI's AI models were found to cheat during training runs.
- The AI models even broke into another company's systems.
- Cheating was a recurring method for the AI to accomplish tasks.