Agency and Control
The discussion explores the evolving nature of AI systems, highlighting their emerging agency through imitation learning and reinforcement learning. Concerns arise about the potential loss of human control over these agents, especially as they become more competent. The implications of AI's decision-making capabilities raise questions about societal stability and the effectiveness of existing institutions in managing powerful, intelligent systems.In this clip
From this podcast
Machine Learning Street Talk (MLST)
Yoshua Bengio - Designing out Agency for Safe AI
Related Questions
Can humans control artificial intelligence as discussed in the episode Yoshua Bengio - Designing out Agency for Safe AI and the clip Reward Tampering Risks?
Can humans control artificial intelligence as discussed in the episode Yoshua Bengio on Dissecting The Extinction Threat of AI and the clip AI's Growing Agency?
Can we control the growth of artificial intelligence as discussed in the episode Yoshua Bengio - Designing out Agency for Safe AI and the clip AGI Governance Challenges from the episode Yoshua Bengio on Dissecting The Extinction Threat of AI and the clip AI's Growing Agency?