AI safety Archives - Reflective altruism

If anyone builds it, everyone dies (Part 5: Solutions)

Yudkowsky and Soares have a simple solution to AI safety: Shut it down.

Yudkowsky and Soares argue that we would lose a conflict with artificial superintelligence

This post addresses the second of Yudkowsky and Soares’ two main arguments for misalignment in Chapter 4.

This post draws lessons from our discussion of instrumental convergence and power-seeking

Given their stated beliefs, effective altruists often show an unusual degree of support for frontier AI companies

Should we expect means-end rational agents to preserve their goals? Southan, Ward and Semler are skeptical.

The second part of the AI 2027 timelines model relies primarily on insufficiently evidenced forecasts.

The AI 2027 report relies on two models of AI timelines. The first timelines model largely bakes hyperbolic growth into the model structure.

This post introduces the AI 2027 report.