AI Experts Say They Need Better Explanations for Everyone

A prominent AI safety researcher is calling for more basic, clear explanations of AI risks. He points out that many existing explainers are filled with jargon, confusing arguments, or are simply missing. For example, there's no good long-form article explaining exactly how an AI could cause a global catastrophe. This lack of clarity leaves the public and even some programmers in the dark about real dangers.

The researcher shares his own experience: he wrote an explainer on AI doom and another on a technical concept called 'neuralese' (a way AI might communicate internally). He found that even top search results were unhelpful. He believes many people underestimate how valuable it is to write simple, accessible pieces. Even experts with busy schedules should spend time on this, and newcomers can contribute by tackling different parts of the information pipeline.

Some worry that low-quality explanations might drown out good ones, but he disagrees. He thinks more explanations—across text, video, and podcasts—are better. The main bottleneck isn't that good explainers don't surface; it's that there aren't enough good ones to begin with. He urges anyone who can write clearly to try, even if they're not a top expert.

The takeaway: if you've ever been confused about AI risks, you're not alone. The solution is more people writing plain-English guides. This could help everyone—from policymakers to everyday workers—understand what's at stake and what can be done.