what will Trump do about AI?

Quick Plug: I’m running a 4-week live course to help you build an AI workflow that actually serves your goals! More details after the main newsletter. ⬇️

I’d imagine most of you would assume this, but in case you didn’t - I am far from thrilled with the results of the latest U.S. Presidential Election. This is for a variety of reasons — most of them having to do with being a queer black woman of color, but I won’t really get into that here.

What I do want to talk about is what AI policy might look like under a second Trump administration. And while I’m not in the prediction business, I do think we have enough information to start making some educated guesses.

In researching for this newsletter, I was surprised to learn that Trump is actually the first president to formally create and launch AI policy initiatives. Obama did have a few initiatives that indirectly included AI (Big Data Initiative, U.S. National Strategy for Advanced Manufacturing, etc.) and emphasized structured regulation/governance, which laid the groundwork for the next administration.

Fun Fact: I’ve only been eligible to vote since I turned 18 in 2015, so Trump has been on the ballot for every election I’ve ever voted in.

Trump (Part 1)

Trump promptly overturned that groundwork when his administration launched the American AI Initiative in 2019, marking the U.S.’s first official step into AI governance. Said “governance” largely boiled down to being very hands-off. The order emphasized keeping the U.S. ahead in the global AI race, particularly against China, by encouraging innovation without putting many rules in place. It also included workforce development and educational initiatives (including support for community colleges and HBCUs), as well as calls for significant investments in non-defense AI R&D. However, while national security applications, like AI in defense and surveillance, got plenty of attention, ethical concerns—bias, transparency, accountability—largely did not.

Let’s Take A Commercial Break: Joe Biden

Biden’s approach to AI marked a significant departure from Trump. His administration’s 2023 executive order introduced accountability measures like public inventories of federal AI use, mandatory safety testing, and anti-bias guidelines. It also included a large AI talent initiative aimed at training federal employees to manage AI responsibly and providing up-skilling opportunities for workers whose jobs might be threatened by further AI development/integration.

This order went a pretty long way towards creating federal government standards for the use and development of AI. However, it did little to curb the private sector, which continued with the “move fast and break things” approach. In fairness, it had little ability to — the ability to regulate Big Tech falls into the hands of Congress, which has been famously good at regulating the private sector lately.

And We’re Back

Fast forward to 2024, and Trump has been re-elected. With that will almost definitely come some shifts in AI policy. He has already committed to rolling back Biden’s 2023 Executive Order, and is expected to adopt a more laissez-faire approach, emphasizing rapid innovation and reduced regulatory oversight, similar to his approach during his first term.

Yes, this could lead to accelerated deployment of AI technologies across various sectors, including healthcare, defense, and advanced manufacturing. And that may lead to attempts to responsibly develop and regulate AI systems being totally steamrolled, resulting in “advanced” technologies that harm people and/or propagate misinformation.

The Elephant DOGE in the Room

sigh

Trump also plans to create the Department of Government Efficiency, otherwise known as DOGE.

Yes, like dogecoin.

This initiative is pitched as a cost-cutting, efficiency-boosting marvel, tasked with streamlining federal operations using AI. Elon Musk and Vivek Ramaswamy are expected to run this ship, which I’m sure will go well based on Elon’s history of “streamlining” Twitter for “efficiency.” It’s also a presidential advisory commission, not an actual federal agency, because only Congress can make those.

If you’d like to get in on that, DOGE is hiring. “High IQ” people willing to work at least 80 hours per week are encouraged to submit their CVs by DMing the DOGE account on Twitter.

Except the DOGE account has Twitter Premium, so you also have to buy Twitter Premium if you want to submit your application, because Elon has to replace that advertiser income somehow.

Anyway.

On top of the Trump administrations’s broader plans, this focus on budget cuts could gut federal research funding for AI. Agencies like the National Science Foundation and the Department of Energy play a critical role in advancing responsible AI development. Slashing their budgets doesn’t make AI innovation “efficient”; it just makes it shortsighted.

DOGE could also undermine talent development efforts. Biden’s workforce upskilling initiatives aimed to prepare the government for AI’s economic impact, but a department obsessed with efficiency might see those programs as expendable. Less training means less oversight, and suddenly you’ve got AI systems running major programs without anyone in the room who understands how they work.

Finally, there’s the deregulation problem. Trump’s efficiency-first mindset leaves little room for ethical guardrails, which means DOGE could roll out AI tools in federal benefits, enforcement, or eligibility systems without adequate testing or accountability. Imagine an algorithm determining your eligibility for unemployment benefits—or worse, denying them—based on flawed or biased data.

In other words, you have nothing to worry about and everything will be fine. 😊

🚨 Something New: Make AI Work For You LIVE🚨

I'm excited to share that I've opened founding member access to my new program "Make AI Work For You" - a 4-week implementation program to help you build an AI workflow that actually serves your goals.

After years of testing hundreds of tools (so you don't have to), I'm sharing my complete system for building AI workflows that work. No more wasting time on endless free trials or money on tools you barely use.

Founding member rate ($147 instead of $347) ends tomorrow at midnight EST.

Learn more and join the next cohort here!

I’d love to work with you!

The Last Loop

this whole newsletter was about news, so we’re skipping the news section 😅