September 14, 2026

3.5 Minute Read

See how the latest AI and digital advocacy tools can transform your work.

THIS WEEK'S EDITION

The Most Important Thing AI Can Do In Public Affairs

In 1955, the Israeli army handed its recruit interviews to a 22-year-old psychologist named Daniel Kahneman. What he found there is still the best explanation anyone has written of what AI is actually for. In today's short read, we show you the most powerful way to apply AI in public affairs.

Delay Your Intuition

The army's interviews weren't working. Interviewers met a recruit, formed a global impression, and graded the impression. It barely predicted who would succeed in combat.

Kahneman scrapped it. Six traits. Factual questions about one trait at a time. Score it, write it down, move on. No global impressions allowed.

The interviewers hated it. In a precursor to the AI age, one told him he was turning them into robots. So Kahneman offered a compromise: finish all six ratings, then close your eyes, picture the recruit as a soldier, and give him a gut score of 1 to 5.

The structured interview beat the old one badly. But the surprise was the gut score. Thrown in to keep the peace, it predicted performance as well as the six ratings combined — and added information the ratings missed. Kahneman gave it equal weight. The Israeli military was still telling interviewers to close their eyes fifty years later.

The whole lesson reduces to one instruction from Kahneman: "delay your intuition, understand all of the data, and then . . . close your eyes."

The Nobel Committee Agreed

That assignment set the direction of Kahneman's career. Five decades later he won the Nobel Prize in economics for the body of work it launched — how people actually make judgments under uncertainty — and wrote the bestseller Thinking, Fast and Slow.

The finding has held up everywhere it's been tested. A meta-analysis of 136 studies found structured, data-driven prediction substantially more accurate than expert judgment. Experts beat the data in fewer than one case in six. The pattern held across medicine, hiring, parole, and academic admissions — and it did not improve with the expert's years of experience.

Fast Thinking Is Great, Until It Isn't

You make thousands of judgments a day. What to wear, which email to answer first, which meeting to take, whether to trust the person across the table. Nearly all of them run on what Kahneman called System 1 — fast, automatic, effortless pattern-matching — and necessarily so. Nobody can deliberate their way through every aspect of a Tuesday.

The trouble starts when a high-stakes decision gets processed on the same circuit as a lunch order.

Consider how decisions actually get made in public affairs. A chief of staff was warm at a reception. A member "feels gettable." A district "seems hot." We form the impression in the first five seconds, spend the next six months confirming it, and call the result experience.

That isn't laziness. There has never been a way to do otherwise.

Until now.

This is where AI can change everything. Nobody can read every floor speech, every committee transcript, every local story, and every post from every influencer, or the countless other sources relevant to deciding who to target, or what to say to them.

But an AI can.

What This Means For Public Affairs

The most important thing AI does in public affairs is play the role of those six ratings, but multiplied many orders of magnitude over. High-ranking elected officials typically have decades of public record. No team could ever make sense of all that information on its own. With AI, it can.

The palpable anxiety in the economy is that AI might replace human judgment. Kahneman's life work shows the opposite. His interviewers didn't become robots. They became better interviewers, because their intuition finally had something honest to work with. No amount of computing power can replace skilled political judgment. But that's a false choice. What AI can do is guarantee that when you make the call, you're making it on the full record instead of the few sources that are readily available to you.

The teams that win the next decade won't be the ones that automate their judgment. They'll be the ones who earned the right to trust it.

Three Ways To Operationalize This Email

🔱 Map the influencers instead of guessing at them. Every decision-maker has an influence orbit. See the actual names in that orbit and layer your skilled political judgment in selecting your audience before beginning any campaign.

🔱 Score the message before you deliver it. A properly trained AI can show you the words most likely to move your decision-maker. Test arguments against the record, not against the room.

🔱 Close your eyes. After the data is in, sit with it . . . and then . . . close your eyes.

🔱 If you're enjoying this content, please consider forwarding this email to a colleague or friend.

🔱 If you're not already a subscriber, please sign up here to stay up to date on the latest developments in political technology.

🔱 If you have suggestions for future newsletters or any other feedback, we'd love to hear from you.