This June, AIC is hosting 100+ events in one week, all built around a single question: what does it mean to be human in the AI era? It's the largest human-centered AI gathering we've ever run, across every chapter, on six continents.

AI-tonomy Summit brings together founders, researchers, investors, enterprise leaders, and AI builders for a full-day gathering on the shift from models to agents. Hosted at Plug and Play Tech Center in Silicon Valley on June 5, the summit explores how autonomous AI systems are moving from conversation to execution — reasoning, coordinating, using tools, and powering real-world workflows.

Across frontier research, enterprise agent adoption, and modern AI infrastructure, AI-tonomy Summit is designed for the people building, funding, deploying, and studying the next generation of agentic systems.

Join 500+ guests and 100+ speakers for high-signal conversations on where AI is going next and what it will take to turn agent demos into scalable products, companies, and infrastructure.

by Noah Frank ⚡️

Molly Kinder of the Brookings Institution published "The Messy Middle" on her Substack last week, adapted from a memo she sent Dwarkesh Patel. Kinder argues that the AI discourse oscillates between two stable framings — Reality 1, today's mostly-intact labor market, and Reality 3, the post-AGI world of abundance whose central question is redistribution of surplus — and skips over Reality 2, the long stretch of disruption between them, which she argues is the period where the most consequential political economy questions of our generation will get decided. Kinder builds her case using two stories of a nearly-fifty former USAID official looking at a sixty percent pay cut to retrain as a teacher, and an early-sixties semiconductor engineer who drove her Uber to Menlo Park and now watches Waymo come for the fallback job. If interested, you can read the full piece here.

One phrase in Kinder's argument, hedged appropriately, concedes more than the argument needs — "if cognition itself becomes commoditized." The thing AI is commoditizing is the knowledge work that looks like cognition from the outside: pattern-matching, document review, drafting, the bounded analytical labor at the base of the law-firm pyramid she describes.

But cognition applied through accumulated judgment, intuition about a specific situation, and a feel for which problem is worth solving in the first place is a different element of knowledge work, and that grows more valuable, not less, as the substrate beneath it gets cheap. The scarcity premium that used to attach to knowledge migrates toward judgment. This is a mechanism claim rather than a measurement, the same epistemic status Kinder flags for her own forecast — both of us reason forward from how the technology works, not back from labor market data that doesn't yet exist.

The related pattern underneath Reality 2 admits more measurement, and it is the gap between AI capability and adoption — the part of every technology transition that goes worst when technologists treat human and institutional friction as details.

Take, for instance, the explosion of interest in vibe-coding tools like Claude Code, Cursor, Lovable, and the rest of the "anyone can ship an app" stack. These tools have made coding and creating open to more people than ever, yet the top performers on the Apple App Store still largely mirror what they have for years: TikTok, Instagram, YouTube, and the same rotating handful of games. Distribution, network effects, switching costs, and brand do most of the work holding incumbents in place, and underneath all of it sits the fact that someone, years ago, understood a customer's problem in a way nobody else did and built the rest around that understanding. Where generation is cheap, knowing what to build and for whom becomes the more important skill.

The obvious next question is what falls out of all this for policy. As I used to say working in AI policy on the Hill, sometimes good AI policy is just good policy and doesn't require taking a position on Reality 3 timelines to make people's lives better. The policies that make Reality 2 livable help whether the messy middle lasts 5 years or 25 years. But more on that soon…

Our partners and sponsors get exclusive placements across the newsletter and access to AIC's in-person network — demo nights, dinners, hackathons, and forums across 180+ chapters.

For all inquiries, send us a note at [email protected].

The AI Collective is built by volunteers across 180+ chapters in 40 countries.

Thank you to the thousands of volunteers around the world who make this work possible. We truly could not do this without you.