Americans are worried about A.I. — but they can’t stop using it. Six in 10 U.S. adults report having used an A.I. chatbot, and 85 percent of those users said the chatbots were useful. But almost half of Americans say A.I. will have a “mostly negative” effect on the human race. Seven in 10 are concerned about A.I. putting people out of work permanently and 53 percent say it’s at least somewhat likely that A.I. will “destroy humanity.”
This split — personal adoption and a more global feeling of anxiety — should give both political parties room to develop unique, comprehensive approaches to A.I. But so far, only Republicans have done so. Democrats, at best, have concepts of a plan. And if they want to stay relevant in 2028, they need to plan carefully now.
The Republican approach, articulated in the White House A.I. Action Plan, is MAGA for A.I. The Trump administration wants to loosen restrictions on domestic A.I. companies, build data center capacity, ensure models don’t generate liberal outputs and set up the United States to outcompete China in the “A.I. race.” It’s vintage Donald Trump: deregulation, pro-corporate policies, social conservatism and hawkishness on China bundled into one tidy package. Some populist conservatives complain Mr. Trump is too friendly to tech giants, but he seems to have set the broad direction for the party.
Democrats are nowhere near as coherent or unified on this issue.
The Biden administration took stabs at building a big-picture A.I. policy. Officials used executive orders to develop standards for labeling A.I.-generated content, to try to prevent A.I.-driven discrimination and to mandate that A.I. companies test their models and share results with the government. Mr. Biden was primarily laying the foundation for a safety-first approach to A.I.
But Mr. Trump rescinded Mr. Biden’s order on A.I. safety, and since then, Democratic action on A.I. has felt piecemeal. Gov. Gavin Newsom of California signed a bill that requires A.I. companies to disclose safety protocols. Senator Mark Kelly of Arizona proposed taxing private revenue from A.I. and channeling it into job retraining. Senator Bernie Sanders of Vermont wants a four-day workweek and an excise tax on A.I. And congressional Democrats want to investigate how A.I.’s voracious appetite for power and water will affect the environment.
These are all mainstream liberal responses to individual aspects of A.I. But they don’t add up to a plan.
Even if there were a perfect, coherent plan on A.I., no Democratic politician has enough clout to unite the party behind it. Until Democrats have a 2028 nominee — that is, for the next two and a half years — Republicans will speak with a single voice on this issue while Democrats stutter.
The good news for Democrats: While they’re stuck in neutral, they can develop multiple plans. And if they play their cards right, they can pick an approach that fits the moment in 2028.
Pro-labor progressives, like Mr. Sanders, would be wise to keep developing a fire-breathing populism on this issue. If an A.I. bubble pops or the technology suddenly vaporizes millions of jobs (or both), Democratic voters will demand a strident, aggressive approach. Progressives will paint Republicans as the pro-robot, anti-jobs party — and they’ll likely grab some swing voters along the way.
“Abundance” liberals — who focus on making it easier to build and expand the supply of infrastructure, goods and the like — also need to keep developing their plans and recruiting politicians. If there is no A.I. bubble and productivity simply shoots through the roof, voters might not be ready to step on the brakes. Instead, Democrats will need to focus on how they can shape A.I. as it grows — emphasizing safety, ensuring that utility bills and carbon emissions don’t soar with energy demand, and planning out how to redistribute, well, abundance.
If A.I. progress falls somewhere in the middle — killing jobs in some sectors, producing wealth and happiness in others, worsening problems like loneliness but failing to threaten the species — Democrats will need to triangulate and improvise. In this scenario, the next Democratic nominee will need to adopt an abundance or even MAGA line in moments when A.I. is improving lives and channeling populist anger when economic, environmental or social issues arise.
He or she will have to weave these decisions together creatively — and most likely come up with a platform that no one can imagine today.
In each of these scenarios, Democrats would benefit from taking a page out of Mr. Biden’s executive orders by addressing emerging noneconomic worries about A.I. Americans fear A.I.-driven job loss — but they also worry that A.I. will lower the quality of education, erode trust in the news media and harm personal relationships. Among chatbot users, 3 percent say they are already addicted, and 5 percent say they worry about becoming addicted. Job loss is a concern, but Americans also feel a broader lack of control over the role of A.I. in their lives.
If Democrats can understand that feeling, they might come up with an answer to MAGA for A.I.
David Byler, a former data journalist at The Washington Post, is a vice president at National Research Group.
Source photographs by prosado, zhen li, ryasick, Sophonnawit Inkaew, SchulteProductions, ayahin and ryasick/Getty Images
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: [email protected].
Follow the New York Times Opinion section on Facebook, Instagram, TikTok, Bluesky, WhatsApp and Threads.
The post A.I. Dominates Our World. Why Don’t Democrats Have a Plan? appeared first on New York Times.
