Democrats Want to Stop AI From Discriminating Against You

New Bill Targets Algorithms in Housing and Hiring

Artificial intelligence is already deciding who gets jobs, apartments, and loans. Sometimes, it’s rejecting people based on their names, their age, or the neighborhood they grew up in.

A group of Democratic lawmakers says that’s a civil rights problem, and they just reintroduced a bill to fix it.

But the timing makes things complicated, because President Trump signed an executive order the same week that could block states from regulating AI at all.

Pressley Leads Push for AI Guardrails

Rep. Ayanna Pressley of Massachusetts joined Sen. Ed Markey and Reps. Yvette Clarke, Pramila Jayapal, and Summer Lee to reintroduce the AI Civil Rights Act on December 2, 2025.

The bill would ban companies from using AI algorithms that discriminate based on race, gender, age, disability, or other protected characteristics.

Pressley, the first Black woman elected to Congress from Massachusetts, said the legislation is needed to protect marginalized communities who bear the burden of biased systems. Twenty-one House Democrats signed on as cosponsors.

The Bill Covers Major Life Decisions

The AI Civil Rights Act would regulate algorithms used in housing, employment, healthcare, education, financial services, insurance, criminal justice, and government benefits.

If a company uses AI to screen tenants, evaluate job applicants, set insurance rates, or approve loans, that system would fall under the law.

The bill’s sponsors say AI has become a $244 billion industry, and the technology now touches nearly every part of American life without meaningful oversight.

Companies Would Have to Test for Bias

Under the bill, AI developers and deployers would be required to conduct pre-deployment evaluations and impact assessments before releasing their systems.

They would also need annual audits to check for discriminatory outcomes. If testing reveals problems, companies would have to fix them.

The goal is to catch bias before it harms people, not after lawsuits pile up.

You Could Appeal to a Human

One key provision gives individuals the right to appeal any algorithmic decision to a human reviewer.

If an AI denies your loan application, rejects your job candidacy, or flags you for extra scrutiny, you could request that a real person look at your case.

Supporters say this restores accountability to a process that has become automated and opaque.

FTC and Private Citizens Could Sue

The Federal Trade Commission, state attorneys general, and private individuals would all have the power to enforce the law.

That last part matters because federal agencies change priorities depending on who controls the White House.

Allowing private lawsuits means people harmed by discriminatory AI could take companies to court themselves, without waiting for regulators to act.

Pressley Points to Inequality in Her District

During a congressional hearing, Pressley described her Massachusetts district as one of the most unequal in the country.

She noted that within three miles of Cambridge, home to Harvard and MIT, lies Roxbury, the most predominantly Black part of her district.

Life expectancy drops by 30 years between those neighborhoods, and median household income falls by $50,000. Pressley called that gap the result of intentional policy and said AI could make it worse if left unchecked.

Studies Show AI Learns From Biased Data

Research backs up those concerns. A 2019 study found that a hospital algorithm used on over 200 million patients gave Black patients lower risk scores than equally sick white patients because it was trained on healthcare spending data that reflected existing disparities.

A 2024 Stanford study found AI chatbots gave worse advice to people with names perceived as associated with Black women. Hiring tools have rejected qualified candidates based on age, disability, and speech patterns.

Courts Say AI Vendors Can Be Sued

Lawsuits against AI hiring tools are growing. In May 2025, a federal judge allowed a class action to proceed against Workday, whose screening software allegedly rejected applicants over 40 at higher rates.

The judge ruled that AI vendors can be held liable for discrimination, writing that drawing a distinction between software decision-makers and human decision-makers would gut anti-discrimination laws.

The ACLU filed a separate complaint against HireVue and Intuit after a deaf, Indigenous applicant was rejected and told to practice active listening.

Trump Moves to Block State AI Rules

Nine days after Democrats reintroduced their bill, President Trump signed an executive order directing the Justice Department to create an AI Litigation Task Force to sue states over their AI regulations.

The order also threatens to withhold federal broadband funding from states with AI laws the administration opposes. Trump said companies need a single national framework, not 50 different state rules.

White House AI czar David Sacks said child safety laws would be exempt from challenges.

Republicans Split Over Deregulation

The executive order exposed divisions within Trump’s own party. Sen. Ted Cruz stood beside the president at the signing ceremony, but Steve Bannon called it a gift to Big Tech.

Florida Gov. Ron DeSantis and Rep. Marjorie Taylor Greene have defended states’ rights to regulate AI. Sen. Ed Markey called the order an early Christmas present for billionaire tech CEOs.

Legal experts say the order may not survive court challenges because only Congress can preempt state laws.

Bill Faces Uphill Battle in Congress

The AI Civil Rights Act has 21 House cosponsors and four Senate cosponsors, all Democrats. With Republicans controlling both chambers, the bill’s chances of passing are slim.

But supporters say they’re building a record for the future.

The Lawyers’ Committee for Civil Rights Under Law, the ACLU, the NAACP, and the Leadership Conference on Civil and Human Rights have all endorsed the legislation.

For now, the fight over who regulates AI is just getting started.

This article was created with AI assistance and human editing.

Read more from this brand:

• Australia Officially Bans Kids From Social Media – and American Parents Are Jealous
• Would you let AI pick your vacation? Google’s betting $700M you will
• Republicans Mad at Trump for Sending $45 Million to Taliban