In keeping with law enforcement’s grand tradition of taking antiquated, invasive, and oppressive technologies, making them digital, and then calling it innovation, police in the U.S. recently combined two existing dystopian technologies in a brand new way to violate civil liberties. A police force in California recently employed the new practice of taking a DNA sample from a crime scene, running this through a service provided by US company Parabon NanoLabs that guesses what the perpetrators face looked like, and plugging this rendered image into face recognition software to build a suspect list.

Parts of this process aren't entirely new. On more than one occasion, police forces have been found to have fed images of celebrities into face recognition software to generate suspect lists. In one case from 2017, the New York Police Department decided its suspect looked like Woody Harrelson and ran the actor’s image through the software to generate hits. Further, software provided by US company Vigilant Solutions enables law enforcement to create “a proxy image from a sketch artist or artist rendering” to enhance images of potential suspects so that face recognition software can match these more accurately.

Since 2014, law enforcement have also sought the assistance of Parabon NanoLabs—a company that alleges it can create an image of the suspect’s face from their DNA. Parabon NanoLabs claim to have built this system by training machine learning models on the DNA data of thousands of volunteers with 3D scans of their faces. It is currently the only company offering phenotyping and only in concert with a forensic genetic genealogy investigation. The process is yet to be independently audited, and scientists have affirmed that predicting face shapes—particularly from DNA samples—is not possible. But this has not stopped law enforcement officers from seeking to use it, or from running these fabricated images through face recognition software.

Simply put: police are using DNA to create a hypothetical and not at all accurate face, then using that face as a clue on which to base investigations into crimes. Not only is this full dice-roll policing, it also threatens the rights, freedom, or even the life of whoever is unlucky enough to look a little bit like that artificial face.

But it gets worse.

In 2020, a detective from the East Bay Regional Park District Police Department in California asked to have a rendered image from Parabon NanoLabs run through face recognition software. This 3D rendering, called a Snapshot Phenotype Report, predicted that—among other attributes—the suspect was male, had brown eyes, and fair skin. Found in police records published by Distributed Denial of Secrets, this appears to be the first reporting of a detective running an algorithmically-generated rendering based on crime-scene DNA through face recognition software. This puts a second layer of speculation between the actual face of the suspect and the product the police are using to guide investigations and make arrests. Not only is the artificial face a guess, now face recognition (a technology known to misidentify people)  will create a “most likely match” for that face.

These technologies, and their reckless use by police forces, are an inherent threat to our individual privacy, free expression, information security, and social justice. Face recognition tech alone has an egregious history of misidentifying people of color, especially Black women, as well as failing to correctly identify trans and nonbinary people. The algorithms are not always reliable, and even if the technology somehow had 100% accuracy, it would still be an unacceptable tool of invasive surveillance capable of identifying and tracking people on a massive scale. Combining this with fabricated 3D renderings from crime-scene DNA exponentially increases the likelihood of false arrests, and exacerbates existing harms on communities that are already disproportionately over-surveilled by face recognition technology and discriminatory policing. 

There are no federal rules that prohibit police forces from undertaking these actions. And despite the detective’s request violating Parabon NanoLabs’ terms of service, there is seemingly no way to ensure compliance. Pulling together criteria like skin tone, hair color, and gender does not give an accurate face of a suspect, and deploying these untested algorithms without any oversight places people at risk of being a suspect for a crime they didn’t commit. In one case from Canada, Edmonton Police Service issued an apology over its failure to balance the harms to the Black community with the potential investigative value after using Parabon’s DNA phenotyping services to identify a suspect.

EFF continues to call for a complete ban on government use of face recognition—because otherwise these are the results. How much more evidence do law markers need that police cannot be trusted with this dangerous technology? How many more people need to be falsely arrested and how many more reckless schemes like this one need to be perpetrated before legislators realize this is not a sustainable method of law enforcement? Cities across the United States have already taken the step to ban government use of this technology, and Montana has specifically recognized a privacy interest in phenotype data. Other cities and states need to catch up or Congress needs to act before more people are hurt and our rights are trampled. 

In early October, a bad actor claimed they were selling account details from the genetic testing service, 23andMe, which included alleged data of one million users of Ashkenazi Jewish descent and another 100,000 users of Chinese descent. By mid-October this expanded out to another four million more general accounts. The data includes display name, birth year, sex, and some details about genetic ancestry results, but no genetic data. There's nothing you can do if your data was already accessed, but it's a good time to reconsider how you're using the service to begin with. 

What Happened

In a blog post, 23andMe claims the bad actors accessed the accounts through "credential stuffing:" the practice of using one set of leaked usernames and passwords from a previous data breach on another website in hopes that people have reused passwords. 

Details about any specific accounts affected are still scant, but we do know some broad strokes. TechCrunch found the data may have been first leaked back in August when a bad actor posted on a hacking forum that they'd accessed 300 terabytes of stolen 23andMe user data. At the time, not much was made of the supposed breach, but then in early October a bad actor posted a data sample on a different forum claiming that the full set of data contained 1 million data points about people with Ashkenazi Jewish ancestry. In a statement to The Washington Post a 23andMe representative noted that this "would include people with even 1% Jewish ancestry." Soon after, another post claimed they had data on 100,000 Chinese users. Then, on October 18, yet another dataset showed up on the same forum that included four million users, with the poster claiming it included data from "the wealthiest people living in the U.S. and Western Europe on this list." 

23andMe suggests that the bad actors compiled the data from accounts using the optional "DNA Relatives" feature, which allows 23andMe users to automatically share data with others on the platform who they may be relatives with. 

Basically, it appears an attacker took username and password combinations from previous breaches and tried those combinations to see if they worked on 23andMe accounts. When logins worked, they scraped all the information they could, including all the shared data about relatives if both the relatives and the original account opted into the DNA Relatives feature.

That's all we know right now. 23andMe says it will continue updating its blog post here with new information as it has it.

Why It Matters

Genetic information is an important tool in testing for disease markers and researching family history, but there are no federal laws that clearly protect users of online genetic testing sites like 23andMe and Ancestry.com. The ability to research family history and disease risk shouldn’t carry the risk that our data will be accessible in data breaches, through scraped accounts, by law enforcement, insurers, or in other ways we can't foresee. 

It's still unclear if the data is deliberately targeting the Ashkenazi Jewish population or if it's a tasteless way to draw attention to the data sale, but the fact the data can be used to target ethnic groups is an unsettling use. 23andMe pitches "DNA Relatives" almost like a social network, and a fun way to find a second cousin or two. There are some privacy guardrails on using the feature, like the option to hide your full name, but with a potentially full family tree otherwise available an individual's privacy choices here may not be that protective. 

23andme is generally one of the better actors in this space. They require an individualized warrant for police access to their data, don't allow direct access to all data (unlike GEDmatch and FTDNA), and push back on overbroad warrants. But putting the burden on its customers to use unique passwords and to opt into—instead of requiring—account protection features like two-factor authentication is an unfortunate look for a company that handles sensitive data. 

Reusing passwords is a common practice, but instead of blaming its customers, 23andMe should be doing more to make its default protections stronger. Features like requiring two-factor authentication and frequent privacy check-up reminders, like those offered by most social networks these days, could go a long way to help users reconsider and better understand their privacy.

How to Best Protect Your Account

If your data is included in this stolen data set, there's not much you can do to get your data back, nor is there a way to search through it to see if your information is included. But you should log into your 23andMe account to make some changes to your security and privacy settings to protect against any issues in the future:

• 23andMe is currently requiring all users to change their passwords. When you create your new one, be sure to use a unique password. A password manager can help make this easier. A password manager can also usually tell you if previously used passwords of yours have been found in a breach, but in either case you should create a unique password for different sites.
• Enable two-factor authentication on your 23andMe account by following the directions here. This makes it so in order to log into your account, you'll need to provide not only your username and password, but also a second factor, in this case a code from an two-factor authentication app like Authy or Google Authenticator.
• Change your display name in DNA Relatives so it's just your initials, or consider disabling this feature entirely if you don't use it. 

Taking these steps may not protect other unforeseen privacy invasions, but it can at least better protect it from the rest of the potential issues we know exist today.

How to Download and Delete Your Data

If this situation makes you uneasy with your data being on the platform, or you've already gotten out of it what you wanted, then you may want to delete your account. But before you do so, consider downloading the data for your own records. To download your data:

1. Log into your 23andMe account and click your username, then "Settings." 
2. Scroll down to the bottom where it says "23andMe Data" and click "View."
3. Here, you'll find the option to download various parts of your 23andMe data. The most important ones to consider are:
1. The "Reports Summary" includes details like the "Wellness Reports," "Ancestry Reports," and "Traits Reports."
2. The "Ancestry Composition Raw Data" the company's interpretation of your raw genetic data.
3. If you were using the DNA Relatives feature, the "Family Tree Data" includes all the information about your relatives. Based on the descriptions of the data we've seen, this sounds like the data the bad actors collected.
4. You can also download the "Raw data," which is the uninterpreted version of your DNA. 

There are other types of data you can download on this page, though much of it will not be of use to you without special software. But there's no harm in downloading everything.

Once you have that data downloaded, follow the company's guide for deleting your account. The button to start the process is located on the bottom of the same account page where you downloaded data.

Our DNA contains our entire genetic makeup. It can reveal where our ancestors came from, who we are related to, our physical characteristics, and whether we are likely to get genetically determined diseases. This incident is an example of why this matters, and how certain features that may seem useful in the moment can be weaponized in novel ways. For more information about genetic privacy, see our Genetic Information Privacy legal overview, and other Health Privacy-related topics on our blog.

The FBI recently released its proposed budget for 2024, and its request for a massive increase in funding for its DNA database should concern us all. The FBI is asking for an additional $53 million in funding to aid in the collection, organization, and maintenance of its Combined DNA Index System (CODIS) database in the wake of a 2020 Trump Administration rule that requires the Department of Homeland Security to collect DNA from anyone in immigration detention. The database approximately houses the genetic information on over 21 million people, adding an average of 92,000 DNA samples a month in the last year alone–over 10 times the historical sample volume. The FBI’s increased budget request demonstrates that the federal government has, in fact, made good on its projection of collecting over 750,000 new samples annually from immigrant detainees for CODIS. This type of forcible DNA collection and long-term hoarding of genetic identifiers not only erodes civil liberties by exposing individuals to unnecessary and unwarranted government scrutiny, but it also demonstrates the government’s willingness to weaponize biometrics in order to surveil vulnerable communities.

After the Supreme Court’s decision in Maryland v. King (2013), which upheld a Maryland statute to collect DNA from individuals arrested for a violent felony offense, states have rapidly expanded DNA collection to encompass more and more offenses—even when DNA is not implicated in the nature of the offense. For example, in Virginia, the ACLU and other advocates fought against a bill that would have added obstruction of justice and shoplifting as offenses for which DNA could be collected. The federal government’s expansion of DNA collection from all immigrant detainees is the most drastic effort to vacuum up as much genetic information as possible, based on false assumptions linking crime to immigration status despite ample evidence to the contrary.

As we’ve previously cautioned, this DNA collection has serious consequences. Studies have shown that increasing the number of profiles in DNA databases doesn’t solve more crimes. A 2010 RAND report instead stated that the ability of police to solve crimes using DNA is “more strongly related to the number of crime-scene samples than to the number of offender profiles in the database.” Moreover, inclusion in a DNA database increases the likelihood that an innocent person will be implicated in a crime. 

Lastly, this increased DNA collection exacerbates the existing racial disparities in our criminal justice system by disproportionately impacting communities of color. Black and Latino men are already overrepresented in DNA databases. Adding nearly a million new profiles of immigrant detainees annually—who are almost entirely people of color, and the vast majority of whom are Latine—will further skew the 21 million profiles already in CODIS.

We are all at risk when the government increases its infrastructure and capacity for collecting and storing vast quantities of invasive data. With the resources to increase the volume of samples collected, and an ever-broadening scope of when and how law enforcement can collect genetic material from people, we are one step closer to a future in which we all are vulnerable to mass biometric surveillance.