4 ms·
Hugging Face: Autonomous AI Agents Should Not Be Developed
- ada1981 2y ago>> Our analysis reveals that risks to people increase with the autonomy of a system: The more control a user cedes to an AI agent, the more risks to people arise. << This seems like a "duh". Risk increases with agency and intelligence.