AI company Anthropic filed for an IPO this week at a possible valuation approaching $1 trillion. OpenAI is expected to follow, with similarly lofty expectations. That’s a tremendous amount of value creation for the owners and employees of those companies. 

The public mood around AI is far less enthusiastic.

Americans would rather have a nuclear power plant in their neighborhood than an AI data center. Nuclear power still carries the stigma of past disasters, but at least people understand the tradeoff: it can provide reliable electricity, stable jobs, and tangible local economic value. 

AI data centers present a very different bargain. They consume huge amounts of power and water, contribute to higher electricity costs, and create relatively few long-term jobs once built. Worse, they power technologies that Sam Altman said less than a year ago were likely to replace 40% of work tasks “in the not very distant future.” This week, he called off the job apocalypse, perhaps realizing that AI in general, and he in particular, are facing a very real image problem.

If the billions in value creation are going to someone else while you are left with higher electric bills and the risk of losing your job, is it really surprising unpopular this is?

It’s gotten to the point where AI has become so overwhelmingly unpopular that it’s becoming a major political issue, in spite of the best efforts of the industry to lobby away legislation.

Bernie Sanders recently proposed the American AI Sovereign Wealth Fund Act, which would require the largest AI companies to pay a one-time 50 percent tax in stock to a federal sovereign wealth fund, giving the public shared upside in their success.

On the right, Senator Josh Hawley has proposed numerous bills to make AI companies legally accountable for their products, with government evaluation of frontier systems and liability when AI causes harm.

Even the Trump administration, which on day one revoked Biden’s AI order and framed earlier safeguards as barriers to American AI leadership, has started to move toward oversight. On June 2, Trump signed an executive order creating a framework under which leading AI developers can give the federal government up to 30 days to review frontier models for cybersecurity risks before release. The executive order makes participation voluntary, but it is a sign of things to come.

The issue is moving beyond Washington. Pope Leo XIV’s recent encyclical on AI warned against the assumption that automation automatically improves work. It raised concerns about deskilling, surveillance, rigid work design, and the loss of worker agency. It’s a moral argument, asking whether these systems are being designed to support human beings or simply to extract more efficiency from them.

The political shift on AI is going to have real consequences for talent acquisition.

The current administration has tried to protect AI companies from regulation, but public opinion is now so negative that some form of regulation feels inevitable. Job loss fears are driving much of that opinion, so hiring will be one of the obvious targets.

Defensibility will matter more.
Talent acquisition teams will need to be able to explain what models are being used, what decisions they influence, what data they rely on, what human review exists, and how the company knows the process is fair. A vague promise that a vendor uses “responsible AI” will not be enough.

Employer experience and brand will have to account for AI anxiety.
Talent acquisition teams that automate large parts of the hiring process without explaining what they are doing will make the process feel cold and unaccountable. But TA departments that use AI while preserving human access, fairness, and transparency can create a trust advantage. They can show candidates that technology is being used to improve the process, not to remove judgment, accountability, or the human relationships at the center of hiring.

TA operations will become even more important.
TA ops is already one of the profession’s growth areas. Decreased public trust and increased regulation will only accelerate that. Governance, workflow design, vendor management, compliance documentation, audit trails, and process controls are not side issues. The companies that use AI well will need people who understand both the technology and the hiring process deeply enough to make it work.

— David

P.S. If you found this newsletter valuable, chances are your colleagues will too. Feel free to forward it along — and if it landed in your inbox by way of a friend, you can subscribe here to get the next one directly.

Sponsored Content

ERE Recruiting Innovation Summit - November 10-11

On November 10-11, join fellow talent acquisition leaders in San Diego for the ERE Recruiting Innovation Summit, a practitioner-led event built around the real work of recruiting today.

You’ll hear from TA leaders sharing what is working, what is not, and what they are still figuring out across AI, hiring strategy, candidate experience, and recruiting operations. And the sessions are only the beginning. Small-group speaker AMAs give attendees direct access to speakers and peers for deeper, practical conversations after each session.

For the first 25 attendees to register, we’re offering a special rate of only $899, nearly half off the full Summit price. There are only a few registrations left at this rate.

Featured Story

AI Can Identify Problems. Recruiters Still Have to Lead.

AI is surfacing hiring risks earlier, from weak candidate flow to process bottlenecks, but the recruiter’s value is in turning those signals into clear decisions. For talent acquisition leaders, Eric Van Duren argues that market intelligence should raise expectations for recruiter judgment, especially in high-stakes fields like healthcare. (ERE)

More Recruiting Insights

Google Search as you know it is over. Google is reshaping Search around AI-generated answers, follow-up questions, interactive results, and background “information agents” rather than ranked links. This is going to fundamentally change how we find information on the internet, and rest assured that Google will find new places for toll booths along the way. (TechCrunch)

Remote Work Leaves Younger Workers Sidelined. New York Fed researchers found that remote work, not generative AI, explains the bulk of the recent rise in unemployment among young college graduates. (New York Fed)

…but remote workers still do not want to return to the office. Readers pushed back on Fidelity’s plan to require 6,200 Boston employees to work in-office five days a week, with an informal poll finding 69% opposed. (Boston.com)

Conferences

ERE Recruiting Innovation Summit

San Diego, CA
November 10-11, 2026

Talent acquisition is moving fast. The best leaders are not just chasing trends. They are comparing notes, testing new approaches, and learning from practitioners who are already deep in the work.

This November, the ERE Recruiting Innovation Summit comes to San Diego for two days of practical insight, honest discussion, and peer-to-peer learning. You’ll hear real examples from recruiting teams tackling today’s biggest questions, from AI and automation to candidate trust, quality of hire, hiring manager alignment, employer brand, and recruiting at scale.

Experience the future of talent acquisition next month. We hope to see you there! (ERE Recruiting Innovation Summit)