Well, the Dems did it! Goodbye to the Wild West of AI and hello deeper AI regulation in the US. New rules will change the way that every industry can deploy and use AI. If you couldn’t tell, this does help us out a little. But it shouldn’t be something that hurts you.
Remember a few weeks ago I said that data centers were gonna cause investors to start really questioning where returns are? Yeah, I wouldn’t have remembered either (check it out here). But it brings up a good point. Do investors realize that the debt bubble is only getting worse, and that they might be told to put a stop to whatever they are building?
Along the Wire
- GPT-6 Cyber pushes businesses to care about security
- Dems win in a landslide, and there are only a few months before regulation becomes major throughout the country
- The DOJ announces the Super Duper Intelligence Task Force
- I lost fantasy football this week to someone who started 0/7 Adonai Mitchell
Dems are up by 40 with 2 minutes left
Democrats won both the House and the Senate on Tuesday, in case you live under a rock. AI and data centers came up all over the campaign trail, and the AP found that most winning candidates for Congress and governor backed stronger rules on data centers and even pausing new ones.
In typical government fashion, it may be a while before anything gets pushed through the floors of the Senate and House (don’t quote me in a few months). Democrats want any federal AI law to set a minimum standard that states can add to, not a limit that overrides them.
So what would a blanket law actually look like? Probably most of the same things the states are already doing: telling people when they’re dealing with AI, keeping a human in the loop on big decisions, and making someone accountable when an AI tool goes rogue. If your business isn’t protected against some of the new laws, you may want to start soon. And if you need a hand, we’re always happy to help.
Current AI Laws
What does this mean for my business though? Well, it just means that it can go in a lot of different ways. Some state proposals have already been pushed like:
- California: if AI makes a big decision about someone instead of a person, you have to tell them up front and let them opt out or ask for a human.
- Colorado: if AI plays a big part in a decision, they must be made aware and you have to let them fix their data, offer a human review, and keep records for three years if they say no.
Both of these kick in on January 1, and they’re only two of the 115 AI laws states passed this year. Big decisions just mean the stuff that actually changes someone’s life: jobs, loans, housing, insurance, health care, and education. So, if you’re using AI to screen tenants, price loans, or sort through job applicants, you’re on the hook, buddy. If you’re just using it to write emails, you’re probably fine.
Heavier blue states are getting the biggest chunk of the pie. They make up a large majority of laws passed, and will most likely continue to get the most focus. Also, I found this cool site that keeps you updated on AI laws (check it out here).
Super Mega Ultra Intelligence Task Force of America
On Thursday, the DOJ launched another task force, promptly named the Super Intelligence Task Force. I thought we all laughed at Super Intelligence and they would have changed it by now? It coordinates federal cases involving AI misuse, takes over the DOJ team that’s been challenging state AI laws, and works with the White House’s own Super Intelligence Force, run by Jay Clayton.
Most of its focus is foreign governments and criminal groups, not the local stuff. But its mandate also covers people and companies using AI for fraud or other illegal conduct. Remember the runaway cow from last week? If your chatbot makes promises to customers that aren’t true, “the AI did it” still isn’t a defense. Now there’s a federal team whose whole job is to find rogue agents and companies using them for nefarious purposes. Need to stay ahead of things like this? Just make some policies and rules around how to control your business’s agents.
Free Things
We built this really cool AI Use Policy generator, and it’s totally free. A written AI use policy is the best thing to build for a team who is just getting started deploying agents. It tells your team which tools are approved, what data stays out of them and who signs off on the important stuff. It’s also the easiest first step toward Wrangl Platinum (subtle ad). Check it out here.
More About Us
Join us next week for more on what’s happening in AI, and reply if there’s something you want us to dig into. Have a great one!
— The Wrangl Team
Sources
- Foley & Lardner: What Democratic control of Congress means for AI regulation
- AP, via Yahoo News: Midterm results signal tougher road ahead for data center projects
- Transparency Coalition: 2026 State AI Legislation Report
- Security Boulevard: US state AI laws in force, what a deployer actually owes
- U.S. Department of Justice: DOJ launches Super Intelligence Task Force
- Government Executive: Justice Department creates new task force to police AI misuse
- AI Weekly: DOJ launches Super Intelligence Task Force to prosecute AI misuse