A Trojan Horse in Congress: 10 Year Moratorium Would Block A.I. Safety Bills

An AI chatbot tells a struggling 14-year-old boy to end his own life—and he complies.

Teen girls are humiliated when their male peers use AI to generate “deepfake” nudes of them, and distribute the images around the school.

Predators use A.I. to exploit children at scale, resulting in 7,000 reports of AI-generated child exploitation in just two years. 

These are only some of the horrifying ways in which AI already facilitates sexual exploitation and harm on a massive scale. In recent years, NCOSE and other experts have been urgently pressing on Congress to get ahead of this problem while they still can. AI is moving fast, and the government must move just as fast to ensure that the explosive industry is held to reasonable safety standards. But now, a Trojan horse has snuck into Congress, which would make this impossible.

Buried deep in a government budget bill, a 10-year moratorium on state AI regulations has been proposed. This means that, for the next decade, states will be barred from passing any laws that promote safety in the AI space.  

This moratorium is unfathomably reckless. It will allow an already dangerous space to become even more dangerous. It will allow the AI industry to completely shirk any responsibility or liability for creating harmful protects. It will allow foreseeable sexual exploitation to go unchecked for at least a decade.

Forty bipartisan state attorneys general, joined by Senators Cantwell and Blackburn, warned Congress that the proposed 10‑year federal moratorium on state AI regulation would silence critical tools states have enacted to curb AI-powered fraud, deepfakes, consumer and artist exploitation, and other harms. They argued that without the ability to enforce laws like Tennessee’s ELVIS Act or Washington’s deepfake and election‑safeguard statutes, Americans would remain vulnerable while waiting for slow‑moving federal rules.

Further, recent research shows that the majority of American voters oppose the AI moratorium by a margin of 3 to 1. Voters’ ppposition to the AI moratorium is bipartisan.

Add your voice to the call to halt this outrageous AI moratorium!

Has Congress Not Learned its Lesson? AI Moratorium is Another Section 230 in the Making

On its face, the AI moratorium is intended to allow the so-called “nascent” AI industry to grow and avoid stifling it with regulations. But Congress has already tried this approach of exempting the tech industry from accountability, in the name of promoting its growth. And it has failed tremendously.

In 1996, Section 230 of the Communications Decency Act came into being. Section 230 gives the tech industry special immunity that has allowed them to escape liability for harmful products. This is a privilege that no other industry enjoys. And the result of it has been thirty years of sexual exploitation and countless other harms proliferating online at unprecedented rates.


Just like the AI moratorium, Section 230 was intended to avoid stifling the burgeoning tech industry. But now, tech is arguably the most powerful industry in the world—and it STILL operates unchecked, free from basic liability that every other industry must contend with.

To this day, children are dying as a result of Section 230 of the Communications Decency Act. We CANNOT afford to repeat and compound this mistake with the AI moratorium.

AI is already taking the world by storm, and its influence is expanding at an astronomical rate. The global AI market is currently valued at approximately $391 billion. It is foolish to pretend that this $391 billion dollar industry is so fragile it can’t withstand the normal legislative process the way all other industries can.

Addressing the Argument that State Regulations Would Result in Contradictory Standards

Proponents of the AI moratorium argue that, because AI is new and legislators do not yet know how best to deal with it, allowing state regulations at this point in time would only result in haphazard and contradictory standards. While this may seem reasonable, it is ultimately misguided.

To begin with, countless industries like telemedicine, ride-sharing, and more thrive despite wide variance in relevant state laws and regulations, proving that innovation doesn’t depend on uniform regulation. For companies, adapting to state-specific rules is simply part of doing business, not a barrier to growth.

Further, there is actually no better way for Congress to learn what kind of AI regulations work well than to allow states to take the first crack at it. It is often said that “states are the laboratories of democracy.” The idea here is that it is a good thing for states to try out various types of legislation because, through that, it will become clear which laws yield the best results. At that point, similar legislation can be adopted at a federal level. So, if Congress wants a 10-year period to learn how best to regulate AI, there is no better way to achieve this than to let 50 different laboratories produce 10 years’ worth of data for examination.

To anyone who thinks this experimental mindset is irresponsible: which is the more dangerous experiment? Allowing states to experiment with different forms of regulation for a billion-dollar industry? Or allowing that billion-dollar industry to experiment on our children, with zero accountability for the harms they cause, when we ALREADY know that the industry is causing rampant sexual exploitation, and even deaths?

The reality is we cannot afford to wait to regulate the AI industry. States are supposed to be able to respond quickly to immediate needs—that is built into the design of America’s federalist system. Federal legislation takes longer to pass because of the bicameral system (two chambers, the House and Senate) and because their decisions impact an entire nation. States are meant to be more agile, moving quickly to address the issues they see their constituents contending with. Handcuffing states with this 10-year moratorium would result in inestimable harm to the American people—and would likely delay the progress of federal legislation as well, since oftentimes momentum at the state level is precisely what spurs federal action.

The conclusion is clear: The AI moratorium would be a disaster for the American people, and most be urgently and firmly opposed.

ACTION: Call on Congress to Oppose the AI Moratorium!

The Numbers

300+

NCOSE leads the Coalition to End Sexual Exploitation with over 300 member organizations.

100+

The National Center on Sexual Exploitation has had over 100 policy victories since 2010. Each victory promotes human dignity above exploitation.

93

NCOSE’s activism campaigns and victories have made headlines around the globe. Averaging 93 mentions per week by media outlets and shows such as Today, CNN, The New York Times, BBC News, USA Today, Fox News and more.

Stories

Survivor Lawsuit Against Twitter Moves to Ninth Circuit Court of Appeals

Survivors’ $12.7M Victory Over Explicit Website a Beacon of Hope for Other Survivors

Instagram Makes Positive Safety Changes via Improved Reporting and Direct Message Tools

Sharing experiences may be a restorative and liberating process. This is a place for those who want to express their story.

Support Dignity

There are more ways that you can support dignity today, through an online gift, taking action, or joining our team.

Defend Human Dignity. Donate Now.

Defend Dignity.
Donate Now.