Deeplinks

Secure messaging platforms, like Signal, WhatsApp, and recently, encrypted RCS, operate on a straightforward assumption: the content at each end of a conversation is private to the participants in the conversation. End-to-end encryption helps provide the mathematical guarantees that the companies who operate these messaging platforms cannot access the contents of messages. But there’s no way to guarantee what happens once the message arrives on a phone. As more devices and services introduce more artificial intelligence (AI) features into messaging apps, that line begins to blur. 

DONATE TO EFF 

ensure ai serves people, not power

When AI features are computed entirely on device, it’s less concerning. Yet sometimes the computing requirements are heavy enough that the computation has to be done on a company server. Tech companies tell us they have a solution for this: trusted execution environments (TEEs). But do server-side TEEs really solve the problem?

TEEs exist to serve many different functions, ranging from digital rights management (DRM) content protections to securely storing information in your phone's mobile wallet, but for our purposes, we’ll be focusing on how tech companies use them for their AI tools. 

The basic idea is straightforward: most consumer devices aren’t powerful enough to handle the sorts of AI features companies want to offer, so sometimes they send data off your device to more powerful cloud servers to do the computing, then display the results on your device. Since your data is leaving your device, there’s a privacy compromise. For example, if you ask for a messaging app to summarize a conversation, it may offload that computing power to a cloud server, sending the entire contents of your messages to the cloud, then back to your phone.

TEEs supposedly offer a way to keep those requests private. There are several implementations out there, like Apple’s Private Cloud Compute, Google’s Private AI Compute, and WhatsApp’s Private Processing. It’s not just the big tech players, we’ve seen chatbots built with TEEs as well.

TEEs can provide more security and privacy than simply running in the clear, but they are fundamentally different from actual encryption or running locally. Despite the promises of some tech companies, they will never be able to match that level of security and privacy. Because of that, a user’s device should never automatically send data to a TEE. Let’s dig through the reasons why.

A TEE is a hardened section of the computer that runs software in a way that’s supposed to be secret even from other processes running on the machine. TEEs also let users check that the code being run is the code that they think is running, and not backdoored code instead, using a process called “attestation.” You may have also heard this referred to as a “secure enclave,” or heard the brand names SGX or TrustZone.

The intention of a cloud-based TEE is simple: a company can run a server in their data center, but still process data that you provide on your behalf without being able to see that information themselves.

In practice, we've seen multiple cracks and hacks every year that show that it is possible to get at that data. That’s because while encryption relies on math, TEEs rely on engineering to provide their security. Standard encryption algorithms are created by years-long processes collaboratively produced by mathematicians around the world and are based on problems that have been studied for decades. The math is reliable, and there is no shortcut to breaking it that would not also upend fundamental understandings of mathematics as a field. 

The collective understanding of every mathematician in the world is that standard encryption algorithms are not breakable to the best of the world’s collective knowledge. No responsible engineer builds a system based on a new encryption method until after it’s been offered up for prodding.

Engineering, on the other hand, doesn’t work like that. Every individual system is the product of a group of engineers who put it out into the world, and each product will have its own quirks and bugs that have to be individually discovered and patched. These bugs are found after the system is built, not before. No one has yet built a system that is unbreakable. On the contrary, there is new research all the time that finds new ways to break into TEE systems. They’re patched as they come up, but they’re unlikely to ever become perfect, and certainly not any time soon. 

TEEs in particular are a hard engineering problem because the encryption key is physically right there on the device. Building a TEE means keeping a key fully separate and inaccessible while it’s on the same physical device as parts of the system that shouldn’t have access to the key.

Many attacks on TEEs involve “side channels.” In a side channel attack, the attacker measures the electrical impulses or other effects to figure out the timing of operations inside the TEE, then uses that to figure out the key being used. Once they have the key, they can read all the data. Compare that to end-to-end encryption, where the key is never on that machine in the first place, so an attacker would have to also run a similar attack on the user’s device.

Companies who turn to TEEs to protect data want to both have the key on the server and have it protected while still performing complex operations like running an LLM, which makes it much more difficult to protect those keys.

That being said, a TEE versus plaintext on a server is the difference between being able to easily read the data and having to do a bunch of specialized work to get at the data. That work often involves accessing the physical machine. This is most relevant for protecting against mass surveillance, and for many people, that might just be enough security.

But that's the core of the problem. “Secure enough for most cases” and “encrypted as in math” are not the same thing, and it’s important not to conflate the two. And services that currently offer “encryption as in math” have a real downgrade in security when they switch to security based on TEEs. 

If you want to dive into the myriad security issues and limitations of TEEs we’ve seen so far, they’re well documented here, here, here, and here.

Sometimes organizations want to offer an LLM that can respond to queries in a private manner. On-device LLMs exist, but they’re limited in size. So, when organizations want to offer the ability to answer queries without being able to see the conversation, they turn to TEEs. That’s a useful way to run a chatbot that’s reasonably private. This is what Apple, Google, WhatsApp, and others are doing.

Why not turn to encryption? After all, LLM inference is just a bunch of math like any other things a computer does. It takes input to a (really big) function and gives an output. We have the math to do that computation in a way that hides the inputs and outputs from the one running the computation, it's just super expensive. It’s called homomorphic encryption, and no one’s figured out how to do it fast enough that it makes sense for this sort of computation.

Instead, the allure of a TEE is that it will run that computation for you inside of a special opaque section of a server. TEE manufacturers try to make it as hard as possible for the person running the TEE to peek inside. But you still have to trust the operator to not put a stethoscope to the box to try to figure out what's happening inside. 

In this case, it’s reasonable to consider these systems “privacy-preserving,” but not “encrypted.” That distinction is important, especially when we talk about how the TEEs interact with secure messaging. When someone using an end-to-end encrypted chat app asks an LLM to summarize, review, or store those messages, the content of those messages is leaving the device and going to an unencrypted third-party server somewhere. That’s a major threat to the privacy of secure chat apps, and one that’s increasingly hard for users to take control of.

The answer to this is going to vary based on an individual’s threat model, but a good rule of thumb is that a user’s device should never automatically send data to a TEE. When the person holding a phone can choose what information is sent, even if it’s a chunk of data like “unread messages,” they have the opportunity to pause and consider if that data might be too sensitive to risk sending.

In contrast, when data is sent automatically, the automatic sending becomes a feature of the system as a whole. If the system was previously end-to-end encrypted, adding automatic exfiltration makes the whole system no longer end-to-end encrypted.

Developers: don’t build systems that automatically send data off a device to a TEE, especially when it’s coming from an app that is otherwise end-to-end encrypted.

Users: if developers ignore us and build that system, turn off any automatic data sending features. Take a second to think about how much you’re willing to risk sending data when you choose to send it off the device.

TEEs are useful for security in a number of circumstances. Your phone likely has a TEE where it keeps the key that encrypts your biometric unlock data and the base of the keychain where passwords are stored. It also enables certain backup systems, like how you can restore a phone with your passcode or restore WhatsApp or Signal backups.

But when we’re talking about cloud processing, it’s important to be clear this isn’t the same as end-to-end encryption and doesn’t offer the same level of privacy. 

Most of our private lives are on our phones and in our messages. We’ve worked for years to secure those messages, with major wins like encrypted RCS, and the continued user experience improvements of Signal and WhatsApp. We’ve even seen real improvements to backup security with features like Advanced Data Protection that bring end-to-end encryption for a variety of data outside of messaging, like notes and photos. 

But as companies roll out AI features that interact with these encrypted services, pulling data off devices and into a cloud-based TEE, they’re eroding the privacy protections of end-to-end encryption and risk causing serious confusion around what data is protected and what isn’t.

With doomsday AI scenarios dominating the news, lawmakers are rightly concerned about reports concerning security breaches at major US AI labs, such as the OpenAI–Hugging Face incident and the many others reported in its aftermath. As they consider potentially regulating frontier AI, they should focus any new legislation on the immediate, demonstrated risks from those incidents. 

Post-incident reports show that the Hugging Face incident could have been mitigated or prevented by following longstanding cybersecurity best practices, like stronger sandboxing and monitoring. Any new legislation should focus on closing gaps in existing law to prevent AI companies from taking unreasonable risks with the public's security.

DONATE TO EFF 

ensure ai serves people, not power

When an AI developer or deployer runs a test or a task that has a high likelihood of causing harm to third parties—for instance, by breaking into someone else's computers—there should be clear minimum safety requirements. Such tests should run in a properly sandboxed test environment, disconnected from other systems, and be monitored and logged. Following these fundamental best practices would have prevented or substantially mitigated all of the incidents at AI labs that we currently know about.

That said, any proposal must be flexible enough to evolve with changing technology. Minimum safety requirements specific only to current AI technologies are likely to become obsolete; legal standards tied to well-established cybersecurity best practices are far more likely to stand the test of time. Tying any new mandates to evidence-backed security protocols also protects the public without impeding future AI development.

Strong legislation should also mandate and fund independent third-party investigations into any serious security incidents that may occur during AI labs’ tests of new tools, and make reports of these investigations available to the public. This important transparency measure would go a long way toward providing public oversight of the industry.

As with any technology regulation, those targeting cybersecurity practices at AI labs must be careful, precise, and practical.

A California law that prohibits teens from receiving recommended social media content from other social media users violates their First Amendment rights, EFF argued this week.

The case, Meta v. Bonta, challenges SB 976, which requires that teen social media users get their parents’ permission before seeing other users’ recommended speech on their social media feeds. The legal challenge to SB 976 has largely centered on how the law violates social media services’ First Amendment rights to curate user-generated content and present it as they see fit.

But the friend-of-the-court brief EFF filed along with the Center for Democracy & Technology and the Wikimedia Foundation shows that the law violates teen users’ First Amendment rights, too.

“SB 976 frustrates young people’s ability to use the internet to its full potential, prohibiting them from relying on tools that disseminate their speech and help them view and interact with other users’ speech,” the brief argues.

Recommendation systems have a dual purpose on social media: they help all users discover speech and content by other users, and to get their own speech in front of a wider audience.

“SB 976 creates significant, constitutionally violative, burdens on young users’ ability to read and comment on the news, discuss politics, find and share art, share their religious beliefs, or even practice their religion with fellow members of their faith,” the brief argues. “There is simply too much content on services for users to sift through manually, and young users may not know what to search for or even how to find content.”

Because SB 976 creates such broad burdens on teens’ ability to distribute and receive speech, it should be struck down on First Amendment grounds. But as EFF’s brief argues, the First Amendment doesn’t stop California and other states from passing laws that help all users, regardless of age, avoid major social media services’ harmful surveillance business models.

“One could imagine a law that required services to minimize the amount of data they collect, or limit using more invasive data analysis practices, such as tracking users across multiple services, analyzing keystrokes, and other surveillance-intensive practices,” the brief argues. “Such restrictions likely would serve the state’s aim of protecting all internet users—including minors—and would be more narrowly tailored to addressing the harms those practices cause than SB 976.”

 The U.S. Court of Appeals for the Ninth Circuit handed internet users and programmers a big win today, by rejecting an attempt to stretch a narrow provision of the Digital Millennium Copyright Act (DMCA) into a new source of copyright liability.  

The case involves Section 1202 of the DMCA, which prohibits intentionally removing copyright management information (CMI) like an author’s name or a copyright notice, from a copyrighted work. Open AI and Microsoft used code from Github as part of the training data for their LLMs, along with billions of other works. A group of anonymous Github contributors sued, alleging the new code coming out of these LLMs was similar to theirs—but with the CMI stripped out.  

The Ninth Circuit correctly agreed with what we said in our brief: removing copyright information from a copyrighted work is fundamentally different from creating a new work that didn't have CMI in the first place. Section 1202 of the Digital Millennium Copyright Act was intended to serve as a backstop for traditional copyrights in the digital age—not to create a new, more expansive right to inhibit otherwise non-infringing uses. 

As we also explained, accepting the Does’ theory would have created a brand-new source of liability for otherwise perfectly lawful activities, undermining creativity and innovation far beyond the specific context of AI development. Copyright holders would be able to file costly lawsuits against all kinds of legitimate users, such as artists making remixes based on older works, teachers adapting works for a classroom presentation, engineers reverse engineering code to understand it better, and search engines that help us all navigate the web. The risks would have fallen especially hard on independent software developers and other small creators. Large companies can afford to litigate these claims in federal court for years, if necessary. But an independent programmer facing massive statutory damages may simply have to settle, even when their underlying use is completely lawful. That’s why EFF fights to make sure courts don’t expand copyright beyond what Congress authorized.  

Copyright law still protects programmers when their work is unlawfully copied. They can still bring copyright infringement claims if someone uses a model to reproduce their code. Additionally, the plaintiffs’ contract claims against the AI companies are still in play. The specific holding here was narrow but important: that the absence of copyright information from a new work does not mean, by itself, that someone illegally removed it.  

That’s the correct result. New technologies will keep raising hard questions about copyright. Courts should answer those questions by applying the rights that Congress actually authorized, not by inventing new rights that could harm expression and lawful use for everyone.  

Additional Reading:  

Courts have for two decades found that linking and embedding someone else’s web content, be it a photo, music, or an article, doesn’t violate copyright law–the entity that controls the server that hosts a copyrighted work, not the user or website that merely directs others to it, is directly liable if the content turns out to be infringing.

News publisher Emmerich Newspapers sought to convince the Fifth Circuit Court of Appeals to chart a new and dangerous course, arguing that an aggregator website that published links to its copyrighted articles was in effect “displaying” them and can be directly liable for infringement. EFF, along with several other public interest organizations and trade associations, filed a brief urging the court to follow multiple other circuits and reject that theory.

Fortunately, the Fifth Circuit Court of Appeals did just that. While it rejected the server test–the rule courts have used to determine copyright liability rests with whoever serves up the content–the court came to the same practical conclusion by focusing on who is responsible for transmitting content. 

Applying that test, the court found that pointing or directing a user’s browser to request and receive the copyright owner’s own copy residing on its computers does not involve transmitting or communicating the content. “Although we take different routes to get there, both the server test and the test we announce end up in a similar place: a website cannot transmit a work that it does not have,” the court said. 

We told the court that accepting Emmerich's theory would make the common act of embedding links a legally fraught activity, one that many websites might be unwilling to risk, which would seriously damage the internet as a tool for creating and disseminating ideas and knowledge,

We applaud the court’s decision–even though it applied a different test, it correctly concluded that a user linking pictures, video, or articles isn’t in charge of transmitting that content to the world. The user doesn’t control what’s located on the other end of the link—that’s up to the person who controls the server.

Emmerich also claimed linking violates the Digital Millennium Copyright Act (DMCA), arguing its URLs were copyright management information (CMI) and when the aggregator displayed Emmerich’s articles under its own URL, it tampered with Emmerich’s CMI, which violates the DMCA. 

Under that logic, unsuspecting internet users could face ruinous legal risk for doing something as simple as using a link shortener, particularly given potential statutory penalties of up to $25,000 per violation.  

In our brief, we told the court that URLs don’t necessarily equate to a copyrighted work or provide sufficient information about the nature of the underlying content, making it highly unlikely that anyone would expect a URL to contain CMI. Quoting EFF’s brief, the court concluded that URLs are first and foremost a locational reference tool and while it may be possible for a URL to contain CMI, the bar to that conclusion is high.

Overall, this was a good and sensible decision that will protect ordinary online expression, communication, and access to knowledge. Hopefully this issue is laid to rest at last.

The Electronic Frontier Foundation (EFF) is honored to announce today that Alexander “amac” Macgillivray — a former White House official who also served in top legal capacities at Twitter and Google — has joined EFF’s Board of Directors. 

Macgillivray served in the Biden Administration as Deputy Assistant to the President and Principal Deputy U.S. Chief Technology Officer in the Office of Science and Technology Policy, and earlier had held a similar position in the Obama Administration. Macgillivray was one of the co-authors of the Biden Administration’s Blueprint for an AI Bill of Rights and oversaw many of the Administration’s AI initiatives, such as organizing its AI CEO convening, leading its working group on federal AI policy, and overseeing the creation of the National AI Research and Development Strategic Plan and National AI Research Resource. 

He was Twitter's General Counsel from 2009 to 2013, leading the Corporate Development, Public Policy, Communications, and Trust & Safety teams. Before that he was Deputy General Counsel at Google from 2003 to 2009, where he created the Product Counsel team.  

“One of the things I am currently focused on is positively impacting AI development,” Macgillivray said. “The EFF is uniquely situated for that purpose because it combines top-notch legal, technical and advocacy staff with a long history of fighting for people’s rights while encouraging the positive development of technology. I’m thrilled to be joining the board.”

Macgillivray joins a dynamic EFF Board led by Board Chair Gigi Sohn and Vice Chair Brian Behlendorf, and including fellow Board Members Erica Astrella, Anil Dash, Sarah Deutsch, Tadayoshi Kohno, Pamela Samuelson, Bruce Schneier, James Vasile, Tarah Wheeler, and Jonathan Zittrain.

“The EFF Board is thrilled to have Alex join our ranks,” Sohn said. “I’ve worked with Alex for over two decades and have always been impressed not only with his intelligence and grace, but also his ability to think outside the box. His deep experience with non-profit boards will be invaluable as EFF enters a new and exciting chapter.”

Macgillivray currently also serves on the boards of The Trust & Safety Foundation, The Trust & Safety Professional Association and Public Resource. He is an affiliate at the Berkman Klein Center for Internet & Society at Harvard University. Macgillivray earned a law degree from Harvard, a bachelor’s degree in Reasoning & Decision Making from Princeton University, and a New Jersey Teaching Certificate. 

“The vanguard leadership of EFF Board members to ensure technology supports rights, justice, freedom, and innovation for all people has never been more critical,” EFF Executive Director Nicole Ozer said. “Many of the threats that once seemed hypothetical are now reality and the work of our EFF community is fundamental to the future of our countries, our livelihoods, and literally our lives. I feel fortunate to have amac join as a Board member as I begin my tenure as Executive Director. His diverse expertise will be invaluable to make sure that EFF is stronger than ever to meet this moment.” 

Members of the Board of Directors ensure the managerial and financial health of the organization.  EFF is the leading nonprofit organization defending civil liberties in the digital world. Learn more about our cutting-edge work on AI issues, and please donate today to help keep us fighting for a brighter digital future.

Donate to EFF

Mass surveillance isn't a joke. But police are treating it like one when using automated license plate reader (ALPR) networks. In our latest EFFector newsletter, we're covering a new EFF report on how officers across the country are routinely logging completely nonsensical "reasons" for their Flock searches, including "LOL," "LMAO," and even (yuck) "Sexy."

JOIN OUR NEWSLETTER

For over 35 years, EFFector has been your guide to understanding the intersection of technology, civil liberties, and the law. This issue covers a settlement enshrining Meta's harmful surveillance into law, states pushing back against ALPR, and how police are turning our privacy into a punchline.

Prefer to listen in? EFFector is now available on all major podcast platforms. This time we're asking EFF's Adam Schwartz what has united people against Flock cameras — and how we can make sure that today's backlash leads to lasting change. You can find the episode and subscribe on your podcast platform of choice:

Want to protect your right to digital privacy? Sign up for EFF's EFFector newsletter for updates, ways to take action, and new merch drops. You can also fuel the fight for privacy and free speech online when you support EFF today!

Since its integration into our digital world, social media has played a pivotal role in youth organizing and social mobilization. Yet, people’s access to these platforms is increasingly coming under threat from courts and legislatures under the guise of protecting young people online—presenting a significant hindrance to youth organizing.  

In a major recent example, Meta settled in a lawsuit with 52 states and territories regarding the use of Instagram and Facebook by young people. The settlement will require Meta, and pressure other non-Meta owned platforms like TikTok and YouTube, to embed age gating practices into every product while also requiring restrictions on the accounts of people under-18, such as a two-hour daily time limit and content restrictions.   

Young people have been using social media for political advocacy and community organizing for more than a decade. From organizing protests speaking out against police brutality, to organizing nationwide school walkouts demanding safety in schools from gun violence, and striking to demand lawmakers take action to protect the climate, social media has become an instrumental tool for youth to both speak out and connect with other young activists.  

Instagram has become especially useful for activism online by young people. The features on the app make it a helpful tool for being able to efficiently and quickly spread awareness, which is especially important when people need to share real-time information. For example, 17-year-old Darnella Frazier’s video on Facebook showed the world the murder of George Floyd. 

The impact of youth activism online is also evident on non-Meta owned platforms, with services like TikTok and YouTube being particularly prevalent spaces for young people to share their stories, build movements, and amplify collective engagement.

However, in a digital world operating under the settlement’s new guidelines, young people risk not being able to read crucial news due to the content being labeled as “age-inappropriate,” which has already happened for teenagers in Australia under its social media ban.  

A two-hour daily time limit and a block on Meta’s apps between midnight and 6am leaves little room for young activists to organize rapid response efforts. Being unable to see likes on a post will make it difficult to gauge the effectiveness of their campaigns.   

Add to this what we already know about Meta’s content policies which claim to “protect children” and keep sites “family-friendly” but instead label content like LGBTQ+ content as “adult” or “harmful,” youth will be left with no choice in what content they see once the ‘age-appropriate’ content filter is turned on by default. One recent report noted that Meta had hidden posts that reference LGBTQ+ hashtags like #lesbian, #bisexual, #gay, #trans, and #queer for users with the sensitive content filter on. This would specifically curtail the efforts of young activists doing work on comprehensive sex education.   

Measures like this are being discussed across the globe, but not all courts have taken such a short-sighted approach. In August, the French Constitutional Council got a lot right in its decision to strike down the country’s legislation banning under-15s from social media for infringing free expression and communication for everyone online, not just young people.  

The French Court also called attention to its infringement on privacy as the legislation would have forced people of all ages to hand over government IDs, face scans, and other sensitive information to prove their age and access online content.  

Requiring this much data from users puts activists in danger of even more surveillance. Meta has already previously complied with demands from law enforcement to hand over the messages of users. The amount of personal information that will be logged and that could be demanded via a warrant from police to stifle or investigate activists’ actions or plans could cause a chilling effect, forcing advocates to pause or terminate their work.  

This is egregious because these systems misidentify or lock out people of color, people with disabilities, and trans or gender-nonconforming individuals whose IDs may not match their chosen name or align with what the system expects them to look like upon verification. And it’s often these communities that benefit from online organizing the most, especially for marginalized youth as social media can often be the only place to organize and build community. 

The settlement generates headlines, but it will not solve the core problem. Instead of tackling Meta’s surveillance capitalism business model that turns all online content into potential profit and centers lining the company’s pockets over protecting the speech and privacy of users, this settlement gives the tech giant an opportunity to carve out a new digital world that prioritizes its own needs, not those of young people.  

As we’ve been calling attention to in other contexts, this will force young people into digital isolation—curtailing vital access to news and resources for health and development. It also completely ignores the calls of youths themselves who favor digital literacy and education over surveillance and government control.   

Young people deserve a better internet than one regulated through panic. They deserve better than the government or Big Tech getting to decide how they use social media and what they can or cannot be exposed to or learn about. They deserve better than having their right to free expression minimized. This must not be lost in the pursuit of building a better and safer online ecosystem and environment.   

The federal government has inserted a provision into a funding deal with the state of California that would make the state abandon its gold standard net neutrality law, broadband affordability laws, and public safety protections. Doing so would be a huge step back for California, and would actually end up being more expensive for Californians in the long run. Tell the governor to reject this provision before accepting these funds from the federal government.

Take Action

Tell the Governor to Stand Up for Net Neutrality, Affordability, and Public Safety

On August 31, the National Telecommunications and Information Administration announced it would be awarding California $1.4 billion to expand broadband connectivity in the state. In that deal is a provision that says that California agrees to not enforce any law, order, or policy that imposes any sort of restriction on internet service providers (ISPs). These ISPs will get awarded the funding in order to connect Californians they have neglected for years. The ban on enforcing our laws would last 14 years. This is disastrous for a lot of reasons. 

First, California is one of the only states with a strong state net neutrality law. Recreating much of the FCC’s Open Internet Order, the law prevents ISPs from blocking, throttling, zero rating, and instituting paid prioritization on internet service. Put another way, the law ensures that users, not companies, decide how they can see on the internet. If California is not allowed to enforce our gold standard law, there will be little stopping ISPs from controlling how everyone experiences the internet. 

Second, California has a number of affordability protections that would also fall under this agreement. For example, when the state approved the merger of Verizon and Frontier earlier this year, it required the new merged company to offer a $20 internet plan to low-income Californians—saving Californians billions of dollars over the next decade. Just this year the California Public Utilities Commission found that the average cost of broadband across four major urban markets (San Mateo, Oakland, Los Angeles, and San Diego) was $51 per month. In 2023, Consumer Reports found that 84% of American consumers pay at least $50 per month, with many paying more. That $30 difference per month—which is likely to actually be more—makes all the difference for low-income Californians. It is how Californians will save billions from this merger requirement. In contrast, $1.4 billion in new connectivity and infrastructure doesn't matter if the most vulnerable Californians cannot afford it. Eviscerations of this and the net neutrality protections will, ultimately, cost Californians more than they will get. 

Third, this deal will impact public safety. The same California net neutrality law which protects consumers also ensures reliable service for first responders during emergencies by banning throttling. In 2018, Verizon throttled, or slowed down, the service of firefighters as they were battling what was, at the time, the largest wildfire in California history. In reaction, fire departments came out in support of what would become California’s net neutrality law. If California cannot enforce its net neutrality law it will leave its first responders in a weaker position as natural disasters only become more intense. 

Most people do not have a choice in ISP as it is. California’s net neutrality law is one of the few things protecting Californians from the whims of these monopolistic giants. Californians should not give up our few hard-won protections in return for a hand out to these behemoths. Tell Governor Newsom to reject this provision before he accepts these funds from the federal government. 

Take Action

Tell the Governor to Stand Up for Net Neutrality, Affordability, and Public Safety

Here's a riddle: Why did a Goshen Police Department officer search 6,474 automated license plate reader (ALPR) networks, representing data from 82,413 cameras, on May 7, 2025? 

If your answer is "I don't know," it turns out you're 100% correct. The officer left the letters "idk" in the search field where cops are supposed to document the reason for the search.

When law enforcement and tech salespeople pitch ALPRs to city councils, they stick to a familiar script. They trumpet the technology, which is often provided by private companies like Flock Safety, Motorola Solutions, or Axon, as an essential tool for solving high-stakes crimes, such as car jacking, kidnapping, or murder.

But when you strip away the carefully curated talking points, the data continues to reveal a different (and frankly, ridiculous) story. An EFF analysis of ALPR search logs from Flock Safety systems shows that officers across the country are spying on drivers for completely nonsensical "reasons." Police are routinely searching the Flock database without providing any legitimate justification, making a mockery of our civil liberties by logging reasons like "LOL" (short for “laugh out loud”), "LMAO" (short for "laughing my ass off"), "sexy," and "idk" (short for “I don’t know”) to access sensitive ALPR location data.

And in some cases, officers are just mashing keyboard buttons rather than articulating the nature of their searches.

Flock Safety claims it has improved its system by requiring officers to select from a dropdown list of crimes before running a search–but that only makes it easier for officers to hide improper searches behind the veneer of uniformity. The system does not require proof that the dropdown reason actually matches the true purpose of the search. 

With no warrant requirements, limited guardrails, and deficient audit processes, ALPR databases have fostered a culture of unrestricted access to everyone’s location information. This culture of abuse has allowed police to treat a mass surveillance network like their own personal search engine, permitting the tracking of the movements of everyday citizens for low-level complaints, personal whims, and sometimes, seemingly, for the lols.

ALPR misuse isn’t a new phenomenon; it has dominated headlines for more than a year. We already know that officers regularly abuse these systems to stalk past and potential romantic partners. We’ve seen ALPRs used to surveil protests, which can chill First Amendment-protected dissent, and seen officers try to use an ALPR system to track down a woman seeking an abortion.

Typically, we learn about these uses from documents called "network audits," which are long spreadsheets that document all the searches that run through an agency's system. It is not unusual for even a small agency to have a record of millions of searches from thousands of external agencies across the United States.

We’ve also uncovered horrific systemic profiling, with more than 80 law enforcement agencies using terms like "roma" and "g*psy" to target ethnic Romani people—often without any mention of a suspected crime. And when police aren’t using ALPRs for stalking or profiling, they routinely use them for extreme low-level investigations: verifying whether a student lives in a specific school zone, running employment background checks, following up on loud music complaints, or targeting a motorcyclist simply for holding a cell phone.

But somehow, it gets worse.

EFF’s analysis of Flock Safety’s ALPR search data obtained through public records requests has uncovered a disturbing trend. In the absence of judicial oversight, officers are inputting ridiculously unserious terms to justify their searches. Here is just a snapshot of what police consider a "reason" to track someone’s vehicle:

Audit logs sample

Button mashing audit logs sample

One of the more alarming discoveries we found in the network audit data is a large number of "reasons" that appear to be nothing more than an officer mashing buttons. These typically involve a nonsensical long string of characters from the same line or area of the keyboard.

For example: 

It's hard to imagine a situation where these characters add up to a legitimate police code. However, it's easy to imagine an officer cutting corners with a text field they know no one is checking, especially if they are accessing the Flock Safety app from their phones while driving. 

When confronted with these flagrantly unserious searches, police departments offered a mix of bureaucratic deflections and excuses. 

In response to EFF’s request for comment, Thornton Police Department (Colo.) claimed the system didn't require officers to select from a defined list at the time, but it does today. They also audited the “driving around being weird” searches, claiming they were all actually for "legitimate public safety purposes." 

Other police departments we reached out to for comment shared the following: 

Other agencies did not respond to EFF’s requests for comment. We will update with responses as they are received.

Under the guise of streamlining audit logs, in late 2025, Flock safety announced that they will be replacing the required, free-text search “reasons” with a pre-populated dropdown menu of generic offense categories. Since this update, officers are no longer required to type out why they are digging through a driver's movement history, and instead can select a pre-packaged option like "Traffic infraction" or “Other” in half a second. 

Replacing the requirement to articulate the reason for the search with one-click searches is a loss for transparency, but also may explain why audit logs including the searches we highlight in this piece significantly decreased since early 2026. 

Entries like these defeat transparency, undermine accountability, and entirely fail to satisfy what many jurisdictions require by law or policy: an actual reason for the search. And this keeps happening because police use ALPRs as a convenient shortcut around constitutional privacy safeguards. 

In other contexts, such as searches of cell phone location information, police have to go to a judge, demonstrate probable cause, and get a search warrant. But because laws and courts have not caught up with the pace of ALPR technology, police do not do the same before searching ALPR databases. Instead, they are given free rein to track a person’s movements without a sliver of judicial oversight.

As we mention in our piece about the use of ALPR surveillance for low-level investigations, if a police chief stood in front of a city council and asked for permission to install hundreds of cameras just so his officers could investigate the high crime of "haha," they would be laughed out of the room. The same could be said if an officer asked a judge to sign a warrant to track someone down for "LOL."

The fact that these searches were not only missed by the agency supervising the officer, but by the often thousands of other agencies whose systems were searched, demonstrates how agencies cannot be trusted to oversee themselves. 

Mass surveillance is incompatible with a free society, and especially so when the people with access to this data are treating it like a joke. This ALPR mass surveillance—the indiscriminate, continuous collection and retention of location data on every driver, regardless of suspicion—should not exist. But because it does, EFF continues to urge courts and state legislatures to immediately step in and impose strict, enforceable restrictions to rein in this abuse. At an absolute minimum, this means mandating rigid data deletion deadlines and an ironclad warrant requirement. 

If police want the power to track a person's movements, they must be required to convince a judge with evidence and probable cause. They should not be able to bypass the Constitution with a search for "haha."