A Berkeley AI professor makes provocative argument for decelerating AI research
UC Berkeley AI professor Anca Dragan argues that AI development should be purposefully slowed to allow society to better understand and adapt to the technology. She suggests that researchers should prioritize safety, transparency, and alignment with human values over racing to build more powerful systems.
Background
The article profiles UC Berkeley AI professor Stuart Russell, a leading figure in the field and co-author of the standard textbook "Artificial Intelligence: A Modern Approach." It highlights his argument that AI development should be deliberately slowed—"decelerated"—to avoid catastrophic risks. This puts him at odds with the prevailing industry ethos of "move fast and break things" and the competitive rush among companies like OpenAI, Google DeepMind, and Microsoft to deploy increasingly powerful AI systems. The key context is the debate between "accelerationists" (who believe rapid AI progress is inevitable and mostly beneficial) and "decelerationists" (who warn that safety precautions are dangerously behind the technology's capabilities). Russell's concern centers on the difficulty of controlling or aligning superintelligent AI with human values once it surpasses human-level intelligence.