The piece argues that the people with the deepest, first-hand knowledge about AI often sound the least certain because honest technical answers require many qualifications. That uncertainty shows up as hedges, corrections and verbal tics during live discussions, and while those signs are evidence of real reasoning they lose out in public debate to polished, overconfident rhetoric from politicians, executives and pundits. That mismatch matters because choices about jobs, regulation, investment and national security are being shaped inside the gap between technical truth and public perception. Formats that let experts reason aloud - long conversations, whiteboard sessions - expose the messy process of thinking and make assumptions visible, but they also risk being dismissed for sounding unpolished.
Practical recommendations follow: compress ideas into ELI5/10/15 versions, pause instead of filling space, give listeners one simple story to carry, state the claim before caveats, use short sentences and layered complexity, and practice writing in public to clarify thought. The author frames these tips through a neurodivergent builder/CEO lens and stresses preserving genuine uncertainty while removing the fog that prevents ideas from traveling. If experts don’t learn to translate calibrated complexity into clear narratives, less accurate but louder voices will dominate consequential decisions about AI.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.