AI is rising faster than any other issue in voter polling, becoming a blank slate for political fear. On Bankless, researcher Jasmine Sun notes politicians like Bernie Sanders are already tying AI to core economic grievances, framing it as an elite project that must be resisted. The potency differs from past tech backlashes because AI drives significant GDP - Anthropic hits a $30 billion annual run rate - and has mass consumer adoption, making the threat visceral.
This populism is fueled by a stark dissonance. Sun reports that while tech CEOs publicly tout ‘small business empowerment,’ they privately describe the median worker as ‘screwed.’ Executives like Anthropic’s Dario Amodei predict AI will break the link between human labor and productivity, a view that lends credibility to public fears. This private pessimism collides with public actions, such as Cloudflare cutting 20% of its workforce while reporting record revenue - a move cited by investor Jason Calacanis as emblematic of the new disruption.
“There is a massive gap between what tech leaders say on the record and what they admit behind closed doors. Sun reports that while executives pivot to talk about 'small business empowerment' during interviews, they privately describe the median person as screwed.”
- Bankless
The scale of potential displacement invites historical comparison. On The Intelligence, Callum Williams argues that rapid, economy-wide job destruction from new technology is unprecedented; even the Industrial Revolution saw British employment nearly triple over a century. He suggests current labor market weakness for new graduates predates ChatGPT. Yet Williams establishes a critical benchmark: if U.S. per-person GDP growth exceeds 2-2.5% annually alongside high profits and broad job losses, it would signal a truly novel AI disruption. Current data doesn’t show this, but the tools for such a decoupling are actively being built.
Those tools are evolving beyond chat. This Week in AI details a shift to persistent, real-time AI that listens and watches, demanding 100x more compute. Nick Harris of Light Matter predicts this will polarize access, with the 1% owning private $10 million data centers for ‘sovereign intelligence.’ The result is a ‘training-layoff loop’ where employees onboard only to have AI learn their roles, leading to purges every 18 months.
“The divide between the haves and have-nots is moving from bank balances to FLOPS. Nick Harris, CEO of Light Matter, argues that the 1% will soon bypass public clouds to build $10 million private data centers in their own backyards.”
- This Week in AI
Sun warns that when people feel they cannot vote on a technology that changes their lives, democratic frustration curdles into radical action. She points to violent attacks on figures like Sam Altman as examples of extreme online nihilism, not a unified movement. The coalition against AI elites is already bipartisan, uniting environmentalists, family-first conservatives, and creatives.
Policy responses are nascent. Sun suggests a grand bargain modeled on historical union bargaining, with longer unemployment insurance, universal healthcare for freelancers, and a shorter workweek to share productivity gains. Without such a contract, the displaced only retain the power to break things. As compute wealth gaps harden and the labor-value link strains, the 2028 election may be less a debate over AI’s promise and more a referendum on who gets left behind.
