Transportation Department plans to write new rules with AI, claims they will be ‘good enough’ – Jalopnik

When it’s not busy creeps help undress childrenSo-called artificial intelligence software can also create flawless Coca-Cola commercials, generate fake YouTube videos about vehicles that don’t exist, and even screw up incredibly basic math. And if you thought to yourself, “Wait, that all sounds bad,” you’d be right. Yet some politicians and top-level executives are obsessed with it, so they continue to force AI on us whether we like it or not. In reality, ProPublica reports The Republican-led Department of Transportation is currently planning to use AI to write transportation regulations.

In December, DOT attorney Daniel Cohen reportedly told employees that AI had the “potential to revolutionize the way we write regulations” and promised a demonstration that would show off “exciting new AI tools available to DOT rule writers to help us do our job better and faster.” Discussions about using AI to write new transportation rules continued after the demonstration was over, through last week. Apparently, Gregory Zerzan, the DOT’s general counsel, wants the agency to become the “tip of the spear” when it comes to federal use of AI and “the first agency fully capable of using AI to make rules.”

You’d think we’d want the rules that planes, trains, and cars have to follow to be written by real people who actually know things, especially since AI’s legal track record full of costly mistakesbut that reportedly doesn’t worry Zerzan. “We don’t need a perfect rule for XYZ. We don’t even need a very good rule for XYZ,” he reportedly said at a meeting, adding: “We want good enough. We’re flooding the zone.”

There’s nothing to see here, folks. Just a bunch of “good enough” rules written by Fancy Autocorrect, intended to govern air travel, crash safety and who knows what else.

These developments have alarmed some at DOT. The agency’s rules touch virtually every aspect of transportation safety, including regulations that keep planes in the air, prevent gas pipelines from exploding and prevent freight trains laced with toxic chemicals from sliding off the tracks. Why, some staffers wondered, would the federal government outsource the writing of such critical standards to an emerging technology that is notorious for making mistakes?

The answer from the plan’s boosters is simple: speed. Writing and reviewing complex federal regulations can take months, sometimes years. But with DOT’s version of Google Gemini, employees could generate a proposed rule in minutes or even seconds, two DOT employees who attended the December demonstration recalled the presenter saying. In any case, most of what’s in the preambles of DOT regulatory documents is just “word salad,” an employee recalled the presenter saying. Google Gemini can do word salad.

In case that wasn’t worrying you enough, Zerzan also reportedly claimed that “it wouldn’t take you more than twenty minutes to get a draft line out of Gemini.” And, as we all know, quantity is much more important than quality when it comes to transportation regulations. Why should concerns about potential problems with a little bit of regulation get in the way of writing as many problems as possible as quickly as possible?

To illustrate this, the presenter asked for a suggestion from the audience on a topic on which DOT might need to write a Notice of Proposed Rulemaking, a public submission outlining an agency’s plans to implement a new regulation or amend an existing regulation. He then plugged the topic keywords into Gemini, which produced a document that looked like a notice of proposed rulemaking. However, it appeared that the actual text of the Code of Federal Regulations was missing, an employee recalled.

The presenter expressed little concern that the AI-produced regulatory documents could contain so-called hallucinations – erroneous text often generated by large language models such as Gemini – according to three attendees.

Of course, the text was missing from the AI-generated concept, but at least it looked official. And it’s not like the text really matters that much when it comes to regulations. It’s more about the general atmosphere anyway, and you can just let people fix any mistakes (if they are still employed and notice them in time). “It seemed like his view on the future of regulation at DOT is that it would be our job to proofread this machine product,” an employee told ProPublica. “He was very excited.”

The December presentation left some DOT employees deeply skeptical. Regulation is complex work, they said, and requires expertise in the subject at hand, as well as in existing laws, regulations and case law. Errors or mistakes in DOT regulations can lead to lawsuits or even injuries and deaths in the transportation system. Some rule writers have decades of experience. But all that seemed to be ignored by the presenter, those present said. “It seems wildly irresponsible,” said one, who like the others requested anonymity because they were not authorized to speak publicly on the matter.

And you know, when you say it like that, it sounds bad. It’s also a step too far for Mike Horton, DOT’s former acting chief artificial intelligence officer, who left his position in August. When Horton spoke to ProPublica, he said the plan was like “having an intern in high school making your rules” and also said those in charge “want to go fast and break things, but going fast and breaking things means people get hurt.” And yes, some of us may die, but as Republicans have shown time and time again, that is a sacrifice they are willing to make.

There’s also a lot more to the original article than would be fair to include here, so go to ProPublica and read the rest of it.