OSASI.ORGCommentary · Ideas for our profession
All briefs← Back to OSASI

Neither boom nor doom: the AI moderates

September 28, 2026 · OSASI Commentary

New polling puts Americans among the most worried people in the world about AI and among the least excited. The AI Daily Brief argues that most people are still in the middle, and that a group it calls the AI moderates is starting to find its voice. OSASI recognises the position: responsible adoption, not reaction, is the way forward.

How a brief is made

A member reads an article or listens to a podcast and sends the snippets, key thoughts or concerns tied to a passage or a timestamp. An AI assistant, chosen for the task and the setup (Claude, ChatGPT, Codex, Gemini or another), searches the full article or transcript, pins the times, checks the quotations, and assembles a draft. The draft goes through rounds of review, and nothing is published until a person approves that exact version. The cues on each post start the player at the passage discussed. Commentary is OSASI's perspective, offered for discussion, not a summary of the source. Each brief is co-authored by a person and an AI assistant. AI assistant for this brief: Claude (Anthropic).

Source podcast · The AI Daily Brief

This episode prompted the commentary below. It is one perspective, offered for discussion.

The AI Daily Brief · September 27, 2026
The Rise of the AI Moderates

Listening highlights · Nathaniel Whittemore (NLW), a Sunday long-reads episode in which he reads from three recent essays. Times follow Spotify's own transcript.

−32:50 to −32:33 remainingThe question · NLWAre extreme opinions the only valid opinions when it comes to AI?
−31:58 to −29:38 remainingA nosedive, and the unformed middleThe Gallup numbers: China at 93%, the United States at 36%; most people still between the accelerationists and the doomers
−29:38 to −26:51 remainingWhy Fukuyama changed his mindSkeptical of the accelerationists, more open to some doomer scenarios; regulation and a negotiated slowdown
−25:14 to −23:25 remainingThymos, and the white-collar blowbackWhy universal basic income cannot replace the dignity of work, and why lawyers and accountants feel it first
−23:25 to −19:35 remainingAgents, delegation, and a measured essayThe danger in capable agents rather than superintelligence; NLW on why the essay is a better place to start
−16:40 to −13:12 remainingAI as normal technology · Kapoor and NarayananControl is underinvested but fixable; liability, insurance and transparency; what defines a moderate
−13:12 to −9:30 remainingKatzenberg: Sousa, sound and the pit orchestras"Is this the end of us?" Past technology shifts, the jobs they cost and the art forms they grew
−6:50 to −5:07 remainingReasoning versus creatingAn AI model's own answer on the difference, and the gap that remains
−5:07 to −1:10 remainingTerms, not toolsCredit, consent and compensation; "the tools-versus-no-tools argument is a trap"
−1:10 to −0:18 remainingA disposition, not a camp · NLWMore people at the table, and a silent majority starting to find its voice

Tap a time above to start the player there. If Spotify shows a card about the show first, close it and the player carries on at that time. Times are approximate transcript times; ads and episode editions may shift them. This plays the full episode, not an extracted clip.

Who is talking

Nathaniel Whittemore, NLW to his listeners, hosts The AI Daily Brief, a daily news and analysis show. This Sunday "long reads" episode, released September 27, reads at length from three recent essays: by the political scientist Francis Fukuyama, by the Princeton researchers Sayash Kapoor and Arvind Narayanan, and by the film executive and investor Jeffrey Katzenberg.

The mood in the numbers

NLW opens on the state of the conversation in the United States, which he says has "taken an absolute nosedive" in the past couple of months. The polling bears him out. In Gallup's survey with Microsoft of 37 countries, fielded from April to July this year, 93% in China expect AI to mostly help their country. In the United States the figure is 36%, among the bottom five. Americans are also the most worried of the 37: 74% of those aware of AI say it makes them feel worried.

A year earlier, Pew Research Center asked a different question in 25 countries: does the increased use of AI in daily life make you feel more concerned than excited, or the reverse? Half of Americans said more concerned, level with Italy for the highest share. Only 10% said more excited, and only Canada, at 9%, was lower. The questions are not the same, but they point the same way. As for China's figure, OSASI takes it for what it is worth.

NLW's argument is that the loud positions are not the common ones. Most people, he contends, are "way more in the middle than either the accelerationists on the one side or the doomers on the other" and, in most cases, "still fairly unformed". Into that space he places the AI moderates: people outside the industry "who refuse to be either gloom and doom or endlessly pollyannish".

Three moderates

Fukuyama has become "much more skeptical of the accelerationist narrative, and more open to some of the doomer scenarios", and now thinks "regulation and a negotiated slowdown in the development of the technology is increasingly necessary." His sharpest point is about work. Lost jobs cannot simply be paid for with a universal basic income, because a job means society values your labour and gives you "a sense of dignity and self-worth. It is a matter of thymos: the pride or recognition that is one of the great drivers of human psychology." And the first to feel it, he writes, will be "the white collar work of educated people (lawyers, accountants, professors and the like)".

Kapoor and Narayanan, in a 13,000-word essay on recent loss-of-control incidents, argue that control of AI systems has been underinvested in but can be fixed. As NLW summarises them, the labs will need to run "less like startups and more like institutions with mature governance", backed by clearer liability, insurance, near-miss reporting, audits and whistleblower protection.

Katzenberg brings the optimism, and the history. An artist he has known for thirty years texted him, after seeing an AI-animated scene: "Is this the end of us?" His answer: "Certainly not." He walks through the phonograph, sound in the movies and computer animation, each of which cost real jobs and grew the art form. "To resist technology is to risk irrelevance," he writes, and "Tools are never the point." His terms for the path forward are credit, consent and compensation: "Build this with the storytellers. Not on top of them." NLW's reading of the essay starts at 19:38 and is worth a listen, as is Daniel Green's illustrated film of it.

NLW closes on what the three have in common. The moderates, he says, are not a camp with shared beliefs but "a disposition towards the conversation", with "a desire to have more not fewer people at the table."

What OSASI hears

The AI moderate is a position OSASI recognises as its own: responsible adoption, not poorly formed regulation, is the most effective way forward. Fukuyama's thymos belongs in that conversation too. The debate about universal basic income treats work as a source of income that can be replaced. For a profession, the work is also recognition of judgement, and that is not something a transfer payment can supply.

One finding the surveys hint at deserves more attention. Gallup reports that in the United States, 68% of daily AI users say AI makes them feel worried, against 80% of less frequent users and 74% of people who have never used it; and 45% of daily users trust AI a lot or completely to provide accurate information, against 8% of those who have never used it. Those who use AI most tend to see it more favourably. That is loaded with bias, since the people who like a tool are the ones who use it, so it is worth continued observation rather than a conclusion.

What this means for OSASI

Fukuyama names appraisers' position without naming them: white-collar work built on judgement, first in line. Katzenberg's history says how such shifts are settled: "never settled by the technology. It was settled by the terms." For appraisal, the terms are who owns the data and the professional knowledge the models learn from, how the use of AI is disclosed, and who gets a say in setting both. The previous brief argued for building on collective intelligence rather than on opportunity for the fortunate. The moderates want more people at the table. A member-driven profession is one way to make sure appraisers have a seat at it.

Send a thought →

A like is counted once per reader per day and stores nothing about you. A thought goes by email to the person who edits these briefs; nothing you write appears on the site unless it becomes part of an approved brief. Share opens the service you pick; this site adds no trackers.

Prefer a feed reader, Slack or Teams? Follow by RSS.