What Others Say About *AI as Normal Technology*

Updated

AI as Normal Technology is a Substack publication by Sayash Kapoor that treats AI as a transformative but ultimately normal technology rather than as superintelligence. The site itself surfaces two third-party endorsements that speak to its credibility: Nathan Lambert of Interconnects AI says the authors "simply don't miss when it comes to covering the most important topics in AI policy," and Ethan Mollick of One Useful Thing calls it "one of the most level-headed observers of AI (even though we don't always agree!)." Together these quotes frame the publication as a reliable, measured voice on AI policy and real-world impact rather than hype.

The endorsements on the page

Source Publication What they say
Nathan Lambert Interconnects AI "Sayash and Arvind simply don't miss when it comes to covering the most important topics in AI policy."
Ethan Mollick One Useful Thing "One of the most level-headed observers of AI (even though we don't always agree!)"

Two things stand out. First, both endorsers are themselves recognized writers on AI, so the praise comes from informed peers rather than anonymous testimonials. Second, Mollick's parenthetical "even though we don't always agree" signals independence — the publication is valued for judgment, not for echoing a party line.

What the reputation implies for readers

  • Policy focus: Lambert's quote specifically credits the publication's coverage of AI policy, so readers interested in governance, regulation, and deployment questions will find it most relevant.
  • Measured framing: "Level-headed" points to analysis that avoids both doom and hype, consistent with the publication's stated premise of AI as "transformative but normal technology, not superintelligence."
  • Scale as a soft signal: The page reports over 89,000 subscribers, which suggests a substantial readership, though subscriber count alone doesn't establish quality.

How to weigh these quotes

Endorsements are curated by the publication, so treat them as a starting point rather than a verdict. A practical way to test the reputation yourself:

  1. Open the site and use "Read without subscribing" to sample posts before committing.
  2. Check whether the topics match your interest — the endorsements emphasize AI policy, so expect that lens.
  3. Compare a post's claims against the primary sources it cites; a "level-headed" reputation is best confirmed by seeing how arguments are evidenced.

If your goal is a steady, policy-oriented read on AI that avoids superintelligence framing, the cited peer assessments support trying it. If you want technical tutorials or product reviews, the endorsements don't speak to that, and you should sample the archive first.