With help from Derek Robertson

Sophisticated, AI-powered facial recognition technology has given law enforcement agencies across the globe almost unthinkably powerful tools for tracking and surveillance. It’s also created intense legal and regulatory headaches, with the European Union considering an outright ban on those tools and many U.S. cities following suit.

But the lure of its power — especially as AI-powered tools seemingly transform every workplace — is still proving hard to resist. The latest example is the Transportation Security Administration, which some senators and experts believe is “moving way too fast” to adopt facial recognition technology three months after the agency ignored lawmakers’ calls to pump the brakes on its usage.

In February five Democratic senators aired their qualms in a letter to TSA Administrator David Pekoske, demanding the agency stop testing facial recognition technology at airports across the country, writing that “Surveillance of Americans by the government represents a risk to civil liberties and privacy rights.”

Rather than heed the warning, in April TSA awarded facial recognition company Idemia a $128 million contract to further the use of the company’s software in airports. Twenty-five airports in the United States currently use the software for biometric assessments.

As for the nascent anti-facial recognition caucus in the Senate? They’re still waiting for a response.

“We have not received a response,” Sen. Jeff Merkley’s office said in a statement to Digital Future Daily. “We should be reining in the use of surveillance technology — not expanding it to unsuspecting travelers.”

The letter — led by Merkley (D-Ore.) and signed by Sens. Ed Markey (D-Mass.), Cory Booker (D-N.J.), Bernie Sanders (I-Vt.) and Elizabeth Warren (D-Mass.) — centered on the TSA’s claim that passengers can forgo facial identification scanning, though it’s “unclear how travelers will know that they can ‘opt-out,’” and what consequences there could be for those who do.

They also highlighted concerns about racial discrimination, citing a study by the National Institute of Standards and Technology (NIST) that found Asian American and Black people were up to 100 times more likely to be misidentified than white men by one-to-one facial recognition matching.

Facial recognition technology can sound like an airy, abstract, inevitable threat, like microplastics or programmatic advertising. Here’s what it looks like at an actual airport security line today: A passenger inserts their license into a slot or places their passport photo against a scanner. They look at a screen in front of them, which compares the photo with their face. If all goes well, the passenger is on their way without handing over any identification to TSA agents — who are still present to sign off on the tests.

“Identity management is a central element to security screening,” agency spokesperson R. Carter Langston said in a statement. “TSA is grounding its exploration of biometric solutions in rigorous scientific study and analysis, to include alignment with NIST standards.”

The facial screening is entirely voluntary, Langston emphasized, as travelers can tell an officer if they don’t want to participate, and noted that accuracy in biometric testing is a priority for TSA. Agency officials have also explained that early testing results have been positive, and the goal is to speed up the accuracy of identity verification without slowing down passengers moving through airport checkpoints.

Despite that, experts worry the technology is being implemented too quickly, and argue that slowing down the process would improve the system.

Both TSA and Customs and Border Patrol, which also uses facial recognition technology, should “take a pause” and focus on improving the identification system through isolated pilot programs, said Jake Laperruque, deputy director of the Center for Democracy and Technology’s Security and Surveillance Project.

In a statement to DFD, Markey renewed his push to stop facial recognition in airports, saying “TSA must halt any use of facial recognition technology so that travelers can continue to fly without checking their biometric data in alongside their luggage at the front desk.”

Booker followed suit, warning about the threat to Americans’ privacy and taking aim at facial recognition generally, as the software systems “too often… misidentify people of color and lead to arrests for crimes they did not commit.”

That’s been an ongoing problem for Idemia, the company awarded the contract last month. The company and local law enforcement officials are being sued for false arrest and cruel and unusual punishment, among other charges, in the case of a New Jersey man who alleges he was wrongly held for 10 days for a crime he didn’t commit due to a facial recognition mismatch.

A former senior Idemia executive also alleges in an ongoing federal suit that the company violated an agreement it signed with the Committee on Foreign Investment in the United States by compromising Americans’ personally identifying information by allowing foreign software developers to create an app used by U.S. citizens. CFIUS is a federal agency tasked with monitoring foreign investments in the United States.

Albert Fox Cahn, founder and executive director of Surveillance Technology Oversight Project, also noted the French-based company’s connections to foreign adversaries, including possible ties to a Kremlin-connected firm.

Idemia spokespeople did not respond to a request for comment.

While Idemia is front and center in some of the U.S.’s most high-profile cases involving facial recognition, experts argue the problem lies with all facial recognition technology.

“Quite broadly, the overall deployment of facial recognition here should be alarming to everyone,” Cahn said. “Because it’s even more of this incremental creep of surveillance theater that seems poorly designed to actually keep people safe, but it’s really problematic from a privacy perspective.”

As to what can actually be done to curb facial recognition’s imminent spread into public life, experts agreed that legislation focusing on an outright ban of the technology is the clearest path forward. That’s already been introduced this year — but relatively few members of Congress have vocally opposed the use of facial recognition, especially for security purposes, indicating that a bill proposing an outright ban would flounder.

Unless Congress acts, it’s likely facial recognition is here to not only stay, but grow quickly around the country.

Daniel Lippman and Alfred Ng contributed to this report.

As Sam Altman did his time in front of Congress yesterday, another, quieter, but potentially more significant AI hearing was going on just upstairs in the same Senate building.

POLITICO’s Brendan Bordelon reported yesterday evening on a hearing held by Senate Committee on Homeland Security and Governmental Affairs chair Gary Peters (D-Mich.) that addressed the use of AI tools by the federal government itself. (On Monday Peters introduced a bill that would mandate AI training within the federal bureaucracy.)

If it all sounds a little dry, think again: Russell Wald, director of policy for Stanford University’s Institute for Human-Centered Artificial Intelligence and a frequent adviser to national policymakers on AI who attended the hearing, told Brendan “the most substantive stuff actually really happened in here,” and that “What the government sets actually sends a powerful message to the rest of what happens in this particular space.”

Ideas floated included a federal “chief AI officer,” literacy-boosting programs like Peters’ proposed bill, and a National AI Research Resource. — Derek Robertson

What’s a little unsolicited advice between tech-bro political dilettantes?

In POLITICO Magazine this morning, Andrew Yang, a 2020 Democratic presidential contender and all-around genial political gadfly, offered some advice for Republican primary upstart Vivek Ramaswamy. It’s nothing too unexpected: Prepare for debates, spend money, show up everywhere — all informed by Yang’s similar experience as, in his words, a fellow “young… with new ideas, tech-savvy, Asian and really, really good-looking,” gatecrashing pol.

But there’s one area where Yang sees Ramaswamy as able to separate himself from the pack by virtue of his youth and background in the tech and business worlds: On AI. “Everywhere you go, you will find people who are anxious about AI,” Yang writes. “They see the world changing underneath their feet. You understand technology and innovation better than just about anyone else — just like I did in 2020. Tell them that you will see to it that they will be included in the economy of the 21st century if you are president.”

Yang, of course, uses this as an opportunity to recommend that Ramaswamy stump for his pet policy cause of universal basic income. But whether the latter takes up that torch or not, he’s likely to appear a much more credible interlocutor on AI than any of his greybearded political rivals. — Derek Robertson

• Maybe it’s not so good for privacy that we can now pull DNA out of thin air.
• Daron Acemoglu makes his case for how humanity can avoid an AI takeover.
• Chatbots could possibly tweak our political beliefs without us knowing it.
• SoftBank is bringing self-driving buses to Japan.
• AI doesn’t necessarily have to kill us to ruin the world.

Stay in touch with the whole team: Ben Schreckinger ([email protected]); Derek Robertson ([email protected]); Mohar Chatterjee ([email protected]); Steve Heuser ([email protected]); and Benton Ives ([email protected]). Follow us @DigitalFuture on Twitter.

If you’ve had this newsletter forwarded to you, you can sign up and read our mission statement at the links provided.