Rohin Shah comments on [AN #118]: Risks, solutions, and prioritization in a world with many AI systems