Bloomberg Surveillance · Wednesday, September 16, 2026
Professor Stuart Russell argues that AI safety necessitates stringent requirements before market access, similar to aviation safety regulations, rather than relying solely on post-event liability. He points to the Boeing 737 MAX incidents as an example of the dangers of bypassing certification processes, emphasizing the need for defined behavioral red lines for AI systems.
“But the second more important point, from a practical point of view right now, is that we need more than just liability.”
“And to get an airworthiness certification, The company that develops the aircraft has to do enormous amounts of flight testing and structural integrity testing and mathematical analysis of the design of the aircraft and software that runs the aircraft and so on.”
“So we need both. Requirements before a developer has market access, as well as liability after the fact.”
“So the answer, we have this concept of behavioral red lines.”