AI and the Crisis of ‘Classical Liberalism’

Compact’s Substack

SubscribeSign in

AI and the Crisis of ‘Classical Liberalism’<br>What do you do when the Invisible Hand is building a totalitarian one-world government?

Geoff Shullenberger<br>Jun 26, 2026

24

Share

This week, the American Enterprise Institute’s new Council on AI Ethics released a “founding document” outlining its principles and goals in the pages of The New Atlantis. The body’s creation responded to an earlier call by AEI fellow M. Anthony Mills for a “President’s Council on AI,” modeled on the W. Bush-era President’s Council on Bioethics. The Trump White House has not established such an entity, so AEI decided to take the initiative and convene a group of mostly socially conservative thinkers to explore “the deepest questions AI is raising about human meaning and purpose.”<br>The document begins by announcing that it will avoid addressing “practical issues like job loss, bias, and the tradeoff in AI development between speed and safety,” and will eschew regulatory debates in favor of philosophical reflection. Although the rationale offered for this reticence is plausible enough, it is also difficult not to see it as evidence of the emerging tension within the right between religious and social conservatives on one hand and tech accelerationists on the other. In an article for Compact this week, editor-at-large Greg Conti identified this emerging “cleavage” around AI as “the most momentous determinant of the future of the American right wing.”<br>Compact’s Substack is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

Subscribe

The Bush administration’s Council on Bioethics, which like the new AEI council was made up mostly of social and religious conservatives, was distrusted by liberals, who saw it as a means of pursuing a regressive theocratic agenda and stymying projects like stem cell research; Barack Obama disbanded the body soon after assuming the presidency. Today, it is left-wing political figures who are more likely to advocate for moratoria on AI development, while the Trump administration has largely taken a “let it rip” approach, a few deviations notwithstanding. But as I’ve argued previously, progressive AI critics often seem to offer little more than “irritable mental gestures” by way of explaining their stance, eliding any substantive account of the challenges of the new technology. In contrast (as I’ve also said before) religious thinkers—not least Pope Leo—have been more perspicacious.<br>The AEI is known less as a hub of religious or social conservatism than as a stalwart of unwavering laissez-faire advocacy. On this ground, one might find the profound reservations about AI enunciated by the new Council somewhat surprising. Today’s tech rightists, who usually define themselves as free-market absolutists, are wont to dismiss any qualms about the technology as a stalking horse for a “woke” or “communist” agenda. If you fail to revel in the glories of the latest model, for the e/acc’ers on X, you are probably a useful idiot for the totalitarian “safetyist” regime bent on stamping out all innovation.<br>Even if it declines to endorse any particular regulations, the new Council’s findings at minimum suggest that there is sound reason for the sort of misgivings that might justify limiting AI development. At the outset, the document cites the bioethicist Leon Kass’s notion of the “wisdom of repugnance,” which holds that “the gut reactions of ordinary citizens to the transformations they see occurring around them are worthy starting points for ethical reflection.” As noted, today’s visceral anti-AI “gut reactions” are often coming from the left of the spectrum, which tends to translate them into proposals like the moratorium on data centers supported by Bernie Sanders and AOC. Kass influenced the attempt at a federal ban on human cloning two decades ago, and it isn’t hard, reading the AEI document, to see how its arguments could be marshaled in favor of a ban on superintelligence.<br>Clearly, the current AI trajectory is bringing to the fore once again the tensions between those two unsteady legs of the fusionist stool: business and religion. But what is more interesting are the dilemmas AI is creating within the GOP business wing’s standard “classical liberal,” free-market framework. Exhibit A of these dilemmas is the influential work of Dean W. Ball, self-described classical liberal and former Trump administration AI advisor and author of its laissez-faire AI Action Plan. Since leaving the administration, Ball has criticized its confusing and haphazard attempts to rein in the industry, in particular its main bugbear Anthropic. Ball has said that he “opposes literally almost all AI regulation,” but he co-authored a New York Times op-ed earlier this year with his Biden administration counterpart, Ben Buchanan, arguing for “appropriate guardrails,” especially when it came to...