Is Government Use of Facial Recognition Software a Fourth Amendment Violation? Unfortunately for Your Privacy, Probably Not

By: Alex Coplan

Congress has failed to implement uniform guidelines for use of facial recognition (FR) programs. Today, many law enforcement agencies use public photos and videos to aid in criminal investigations. There are concerns that wide-spread public surveillance and use of these programs could violate the Fourth Amendment. However, government observation of people in public generally falls outside the scope of Fourth Amendment protections. Since law enforcement’s use of FR software only enhances public photos and videos, it likely does not count as a “search” under the Fourth Amendment.

Constitutional Implications

Constitutional concerns with FR arise under an individual’s right to privacy. Under the Fourth Amendment’s protection against unreasonable searches and seizures, the question is whether people have an actual expectation of privacy in public, and whether society recognizes this expectation as reasonable. Currently there is little federal framework specifically directed at the use of FR technologies by government and law enforcement entities. Under the third-party exposure doctrine, government observation of people in public is not a “search” for the purposes of the Fourth Amendment. However, as technology improves, so does the ability to survey a person’s public activity. Furthermore, when detailed information is collected regarding a person’s movement, a privacy interest may arise because of the cumulative nature of the information received. Indeed, the Supreme Court in Carpenter v. United States found that long, continuous surveillance using new technologies may implicate privacy interests protected under the Fourth Amendment. Therefore, the constitutionality of using FR technologies may be dependent on how exactly FR services are used.

What Can We Learn from Other 4th Amendment Cases?

The United States does not directly regulate the use of FR technology by either the private sector or government entities. However, a few court cases may shed light on how privacy and the use of electronic surveillance connect.

Government use of technology to monitor public activity may not violate the Fourth Amendment. In United States v. Knotts, the Court addressed whether a GPS tracking device placed in a container and obtained by a suspect exceeded Fourth Amendment limitations. Finding that the device only revealed information that could have been viewed by the general public—like the suspect’s travel on public roads—the Supreme Court held that the use of this technology in this limited capacity did not violate the Fourth Amendment. This holding suggested that the government may use technology to monitor a suspect’s movements in public without a warrant, when that surveillance could be achieved through regular visual surveillance by anyone, not just the police. However, the Knotts court warned that constitutional issues may arise if “twenty-four hours surveillance of any citizen of this country [were] possible.”

In another case that explored GPS motion tracking, concurring Justices in United States v. Jones expressed concerns about extended use of surveillance technology. There, the Court held that placing a tracking device on a suspect’s vehicle constituted a “trespass” prohibited by the Fourth Amendment. In Jones, the FBI placed the tracking device and analyzed the suspects movements for 28 days. In the FR context however, the concurring Justices focused on long-term tracking of a person’s public movements, the information of which could be used for “years into the future.” It was not long before the Court’s concern in Jones—using technology for constant surveillance—was realized.

Long-term surveillance of a suspect may constitute a “search” under the Fourth Amendment. In Carpenter v. United States, the Court held that law enforcement’s use of cell site location data spanning over a 152-day period, which allowed law enforcement to track a suspect’s location, constituted a “search” under the Fourth Amendment. Since this form of surveillance was considered prolonged and continuous, it overcame the presumption of the third-party exposure doctrine that surveillance of public activities does not raise Fourth Amendment issues.

However, the Carpenter court cabined its holding to real-time tracking for periods less than seven days, explicitly stating that the holding was a “narrow” one. While Carpenter and Jones cautioned against long-term tracking of suspects, FR itself may not raise the same exact issues. The use of cell site location information and FR technology is unlikely to be analogous, in that FR technology by itself is not generally used for continuous surveillance. Importantly, FR is not the means of surveillance, but merely enhances photos and videos already taken. In other words, video or photo surveillance in conjunction with FR technology is the concern.

This concern was addressed in Leaders of a Beautiful Struggle v. Baltimore Police Department. In this Fourth Circuit case, the Baltimore Police Department contracted with Persistent Surveillance Systems (PSS) to conduct its Aerial Investigation Program (AIR), which is equipped to fly drones over the city of Baltimore and take video for 40 hours a week. This aerial surveillance technology, combined with BPD’s resources, can give detailed information about an individual’s identity and activity. Regardless, the Fourth Circuit concluded that people do not have a right to avoid being seen while in public and that the AIR program sought to meet serious law enforcement needs with limited privacy invasions.

FR Use Does Not Violate the Fourth Amendment, but Further Regulation is Required

As long as FR only uses public video and photo surveillance, capturing only what the naked eye can see, the use of FR only enhances information already in public view. Additionally, law enforcement’s use of DMV photos or other images held by third parties, combined with FR technology, likely does not violate Fourth Amendment privacy rights. As a result, the third-party exposure doctrine likely applies, as was found in Leaders of a Beautiful Struggle v. Baltimore Police Department. Indeed, although the Court in Carpenter found the prolonged use of cell-site location information a violation of the reasonable expectation of privacy, the purely visual and public nature of the surveillance with regard to FR programs would likely not implicate the Fourth Amendment.

The cases described above demonstrate that the use of FR technology in public likely does not result in an unreasonable search, as long as the surveillance remains in a public place. The use of FR programs in more private settings, however, may lead to privacy implications protected by the Constitution.

As a result, the weight of the problem falls more heavily on the accuracy of FR programs rather than Fourth Amendment issues. If an FR program were to be wrong, an innocent person may be investigated, or even arrested, for a crime they did not commit. This is an abhorrent result for a technology made exclusively for the purpose of making positively identifying criminals easier. Federal legislation is hopefully not too far off. Until FR can be perfected, national regulation is required to ensure the use of such software is not abused.

Post-mortem privacy rights: dying with indignity in the digital age

By: Nicole Buckley

Last Friday night, my roommate struck up a conversation with me as he heated rice on the stovetop. “Did you hear about the murder-suicide in Pennsylvania yesterday? Some guy shot his neighbors, then went back inside his house, grabbed an AR-15, and sprayed their bodies with bullets. A home security system caught all of it.” I paused, then: “Wait. Are you saying you clicked on a video and watched two people die?” Indeed, he had.

Although gruesome content is de jure banned on major social networking sites like Facebook and Twitter, other platforms are much more liberal in what content they permit users to post, view, and share. To that end, “free speech” is a dog whistle for graphic, violent, or hateful content on forum-style websites like Reddit, Gab, and Bitchute. Rather than removing or obscuring graphic content, these “free speech” platforms tend to be breeding grounds for the fetishization of extreme violence. The video my roommate elected to click on is one example among many.

At present, the legal community is laser-focused on so-called “free speech” platforms, Section 230 of the Communications Act of 1934, and election-related disinformation. Admittedly, I am, too. However, questions about post-mortem privacy in the digital age—for now lurking beneath the surface of popular debate—seem primed for the spotlight in years to come. If alternative platforms are positioned as ready and willing to host videos depicting murder, what is the legal community to make of it? How far does Section 230 extend? If users are unable to reach platforms to remove the content, would they be able to bring a tort claim against other users on behalf of the decedent(s) depicted? Some questions are easier to answer than others. The second question is difficult to answer. The last question is likely to produce a simple response: no.

Ironically, even though privacy rights appear to be an invention tailored to curtail the inquisition of technology on various subsections of society, privacy law lags behind advancements in the digital age. Thus, despite public opinion surveys suggesting a general desire to expand privacy rights online, the expansion of those rights has proved difficult. This is largely because an individual’s privacy rights are thought to terminate upon their death. Salting the wound is the common law majority view that a decedent’s estate may not bring an action to protect that decedent’s privacy interests. As such, depictions of gruesome deaths can be extraordinarily difficult to remove from the internet.

The debate around internet-era post-mortem privacy rights has been brewing for over a decade. After a teenager died tragically in a car accident, photographs of her decapitated body were plastered across the internet. Although the decedent’s family pursued legal recourse, eventually settling with the law enforcement agency responsible for leaking the photographs, a quick search on Reddit reveals that the images are still being circulated. Two interesting points emerged from this case. First, the California Court of Appeals never reached the privacy claim on the merits due to the family’s decision to accept and out-of-court settlement. Therefore, although the court appeared willing to entertain a privacy discussion, the privacy issue was not litigated and the law is thus unclear. Second, and prior to settlement, the law enforcement agency relied on the First Amendment as a defense.

The collision of Section 230, the First Amendment, and post-mortem privacy is inevitable. First Amendment jurisprudence, often thought to be in tension with privacy rights, permits platforms that host content depicting graphic content to claim “newsworthiness” as a legal defense. Additionally, Section 230 absolves internet platforms of liability for failing to remove most user-generated content. Not only are decedents disadvantaged in a common law privacy context, but constitutional jurisprudence and federal regulatory law also fail to provide litigants with adequate legal recourse.

Thus, theoretically available avenues to relief appear sealed shut: decedents’ estates cannot easily sue the people who shared content depicting a loved one’s death, nor can they get at the platforms hosting the content. As Danielle Citron Keats points out, the available private responses meant to curb abuse online are wholly inadequate. Yet, all hope is not lost; there is still time to broaden scholarly debate about private rights in the context of the digital age. After all, the common law is not inflexible, and debate around Section 230 is presently reaching a fever pitch. As a new generation of lawyers enter practice, it is imperative we consider online privacy rights against their origin story: a mechanism developed not only to check technology, but to balance its uses with the good of society. If death is to retain dignity in the digital age, private rights of action must be accorded more opportunity to succeed.

Settlements Are No Longer an Option: Utilizing Cameras to Pierce the Shield of Qualified Immunity

Photo by Karolina Grabowska on

By: Marcena Day

Have you seen the video of the Black child killed by the police? You know, the one where the police threw a flashbang into a house?  She was asleep in the living room with her grandmother when her blanket caught fire from the flash bang. The police appear frightened (but courageous enough) to enter the unknown domain—her home. Lifeless, Black flesh lays on the floor. Have you seen it? 

Aiyana Stanley-Jones was seven years old when she was killed by Detroit Police Officer Joseph Weekly during an arrest warrant execution. In May 2010, a TV camera crew was filming the operation for an upcoming episode for a police reality show. Once again, Black Americans and other racial justice allies instinctively demanded justice. The marches and rallies had similar messages following Breonna Taylor’s death in 2020. These demands highlight how Black Americans are historically conditioned, by both the justice system and the political-social sphere, to be subjected to state-sanctioned acts of anti-Black violence. Today, Black Americans have a crucial tool at their disposal: video recordings of their interactions with the police. 

Video recording of police interactions have the power to both  implicate and exonerate officers. But what purpose do these recordings have in an anti-racist criminal justice system? Social scientist James C. Scott, in his book Domination and the Arts of Resistance, contends a significant aspect to maintain relations of domination “consists of the symbolization of domination by demonstrations and enactments of power.” Saidiya Hartman , in her book Scenes of Subjection: Terror, Slavery, and Self-making in Nineteenth-Century,supplements, “[t]hese demonstrations of power consisted of forcing the enslaved to witness the beating, torture, and execution of slaves…such performances confirmed the slaveholder’s dominion and made the captive body the vehicle of the master’s power and truth.” Perhaps viral video recordings of Black Americans being beaten, tortured, and executed only reinforce white supremacy in the United States? 

This article grapples with this question. Part I discusses the public’s right to record the police and how video recordings help plaintiffs navigate 42 U.S.C. § 1983 suits (lawsuits that allow people to sue state actors such as the police for civil rights violations). Part II explores the legal stalemate § 1983 suits face due to qualified immunity. Part III recommends § 1983 claimants to refrain from settling and instead use video recordings to help pierce the shield of qualified immunity. 

Using video recordings of police interactions to prove § 1983 suits

The facts and circumstances are crucial in § 1983 suits, thus video recordings may offer the “truth objectively” that can implicate or exonerate officers when the facts are unclear. 

The Ninth Circuit, in Fordyce v. City of Seattle,  recognized a First Amendment right for individuals to record police and other public officials performing their public functions in public spaces. Armed with this knowledge, Black America frequently records their interactions with the police. Philando Castile’s girlfriend, Diamond Reynolds, recorded his execution on a Facebook live. Dreasjon “Sean” Reed similarly captured his final moments on his Facebook live. Police body cameras and helicopter footage recorded Stephon Clark’s killing in his backyard. Most infamously, police body camera video recorded the 8 minutes and 46 seconds of an American being tortured and killed on the street. In each case, their descendants filed suits alleging excessive force.  

Individuals may sue the police for excessive force under 42 U.S.C. § 1983. In § 1983 suits the officer’s actions are judged by a reasonableness standard “not capable of precise definition or mechanical application.” The application of the reasonableness standard “requires careful attention to the facts and circumstances of each particular case.” Relevant factors to evaluate include “the severity of the crime at issue, whether the suspect poses an immediate threat to the safety of the crime at issue, whether the suspect poses an immediate threat to the safety of the officers or others, and whether he is actively resisting arrest or attempting to evade arrest by flight.”

Criminal Procedure Professor Mary Fan argues, “[f]actual details make all the difference in civil rights suits, and it is these details that are intensely disputed and hard to reconstruct.” Accordingly, video recordings can “offer an important source of evidence to courts and juries wrestling to apply fact-intensive standards.” She suggests there is hope that recordings will present the “truth objectively.” Fan contends, “the power of video is to take the case directly to the people, generating pressure to settle cases outside the formal confines of the courtroom and doctrines such as qualified immunity.” This argument is notable and highly persuasive, but choosing this route comes at a high price in an anti-racist criminal justice system. 

Qualified immunity sets a high bar for justice

Qualified immunity shields police officers “from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.” The plaintiff must allege facts showing a violation of “a statutory or constitutional right that was ‘clearly established’ at the time of the challenged conduct.” To be clearly established, “existing precedentmust have placed the statutory or constitutional [violations] beyond debate.” (emphasis added). To have “existing precedent,” a previous court must enter a final order on the merits. Essentially, plaintiffs go to trial for a court to enter a final order. This creates stare decisis—which governs how courts apply rules to future cases. Yet, even if plaintiffs go to trial, § 1983 suits may be dismissed based on qualified immunity by the judge without her determining whether the conduct alleged violated a statutory or constitutional right.

This poses a unique (and deadly) legal stalemate: there may never be an existing precedent that places the injuries alleged in a § 1983 suit “beyond debate” because of (1) judicial discretion or (2) the plaintiff settled prior to trial. Which begs the question, what is the purpose of video recordings of the police beating, torturing, and executing Black Americans if the court never enters a final order to hold the police accountable? 

Plaintiffs should refrain from accepting settlements to overcome qualified immunity

It is highly unlikely to curb judicial discretion, but plaintiffs can utilize video recordings to demonstrate similar and recurring violations of “a statutory or constitutional right that was ‘clearly established’ at the time of the challenged conduct.” 

Recall Detroit Police Officer Joseph Weekly. He and his team executed a no-knock warrant—a warrant that allows police to enter property without prior notification—in the killing of Aiyana Stanley-Jones. All charges against Weekly were dismissed and the Supreme Court denied hearing an appeal. On April 2, 2019, Weekly returned to active duty on a “restricted” basis. The family of Stanley-Jones agreed to an $8.25 million settlement with the city of Detroit. The story of Stanley-Jones is similar to other Black women killed by the police during a no-knock warrant. Simply put, the fact pattern and underlying statutory or constitutional right violations of Stanley-Jones’ § 1983 suit may transcend beyond her own experience

The shield of qualified immunity can be pierced if litigators collect video recordings of similar fact patterns that suggest the same statutory or constitutional violation thereby creating a “clearly established” constitutional right. Thus, plaintiffs should redirect the power of video recordings away from exacting settlements and toward demonstrating the same statutory or constitutional violation that can be ‘clearly established’. 

For no-knock warrants, litigators can compile video recordings with similar fact patterns and argue the police use of force violates the same searches and seizures rights protected by the Fourth Amendment. But plaintiffs must first reject settlement to weaken the legal stalemate § 1983 poses to help build stare decisis. As Black Americans continue to record their interactions with the police, this application could be broader than no-knock warrants, e.g., Terry stops and frisks as seen in the cases of George Floyd, Philando Castile, Dreasjon “Sean” Reed, and Stephon Clark. Otherwise, the video recordings are merely another video of an American citizen being beaten, tortured, and executed by officers sworn to protect them. 

‘Brexit’ and its Effects on International Trademark Law: Attack of the Clones

Michael W. Meredith

The United Kingdom formally left the European Union on January 31, 2020. Since that time, the country’s legal system has been in an eleven-month ‘transition’ state that has delayed the effects of the country’s departure. However, that transition period came to a close on December 31, 2020 and the New Year will now force both the public and any businesses operating in either the European Union or the United Kingdom to reckon with the full brunt of ‘Brexit.’

The U.K., the E.U., and the International Trademark Register

One of the many legal challenges presented by ‘Brexit’ has to do with a peculiarity in the trademark registration system of the UK vis-à-vis the European Union and the International Register. The International Register is a global trademark protection system created pursuant to the Paris Convention for the Protection of Industrial Property of 1883 (the “Paris Convention”)—a multilateral treaty that permits its signatories to efficiently and cost-effectively apply for, register, and manage their trademarks in multiple jurisdictions across the globe using a single, centralized system. One of the primary benefits of that system is that it allows trademark owners to pursue a single ‘international’ trademark application that, if approved, would provide the applicant with trademark rights in any of the Paris Convention’s member countries designated in the application.

Both the European Union and the United Kingdom are signatories to the Paris Convention, but because an EU-level trademark registration protects the use of a mark in all European Union member states, international applicants will often choose to designate the European Union as a whole, instead of the United Kingdom or any other individual EU member state, in their international applications.

The same is true with respect to trademark owners who have not elected to pursue international trademark registrations, but still wish to do business throughout Europe. Often, they will seek trademark protection with the European Union’s Intellectual Property Office instead of pursuing multiple duplicative applications in each EU member state. These otherwise prudent and cost-effective trademark protection strategies, however, are now being challenged by ‘Brexit,’ as owners of international or EU-level marks are now left to question whether their rights will be protected in the newly-independent UK.

The Effect of the U.K.’s Withdrawal Agreement

The answer to many of their concerns is included in a Withdrawal Agreement that was reached between the EU and the UK on February 1, 2020. The Agreement provides that, for any EU-level trademark registration or international registration designating the EU, the UK will create a duplicate or ‘cloned’ registration in the UK’s national registry, affording registrants with the same trademark rights they would have had in the UK, pre-‘Brexit.’ But this ‘cloning’ process will not come without certain costs to the trademark owner. The newly-created UK registration cannot be managed or maintained through the EU Intellectual Property Office, or the international trademark registry. Indeed, UK trademark attorneys are not even permitted, post-‘Brexit,’ to serve as representatives at the European Union Intellectual Property Office. Instead, ‘cloned’ registration must be managed directly through the United Kingdom’s national office and will require the retention and appointment of a UK-approved trademark attorney or representative.

A ‘cloned’ entry will also not be provided in the UK registry for any international or EU-level applications that were ‘pending’ as of December 31, 2020. As such, should the owner of a pending international application designating the EU hope to receive trademark protection in the UK, they will need to appoint a UK-approved trademark attorney or representative to prepare and file a parallel application with the UK Intellectual Property Office before September 30, 2021, or they will be unable to assert the ‘priority date’—the date that their trademark rights are deemed to begin—provided for in their existing EU or international application.

What’s Next?

An open question is what effect the UK’s exit from the European Union will have on applications or registrations that are subject to a challenge by third parties. EU trademark applications and registrations may be challenged by any third party if: (1) that third party has made relevant use of a similar trademark in the EU prior to the date of a potentially infringing trademark’s application or registration; and/or (2) the registrant has failed to make use of its mark in the EU for a period of five years. The UK’s Withdrawal Agreement provides that even EU trademark registrations that are subject to a third-party challenge will be ‘cloned’ in the UK trademark registry and that UK applications duplicating pending EU applications that have been challenged may be filed before September 30, 2021. But the Agreement is silent with respect to the duplication of pending opposition or cancellation proceedings, suggesting that duplicate challenges will need to be filed by third-parties against the UK ‘clones.’

The UK has indicated that should a ‘cloned’ EU trademark registration be successfully challenged by a third party and removed from the EU registry, the UK ‘clone’ will also be removed. But it is unclear whether that is true or legally permissible if the challenge to the EU trademark is based upon the use or non-use of a trademark only in EU member states, but not the UK, or when multiple grounds for the challenge have been alleged. Prior to ‘Brexit,’ such a challenge would result in the cancellation of an EU-registered trademark but not a UK registration. As such, the effect of a similar challenge to a UK ‘clone’ is not at all clear and the UK Intellectual Property Office has left the majority of these disputes to be resolved according to its own discretion, noting that: “[i]f a third party sends . . .  a cancellation notice [regarding an EU registration]. . . [t]he IPO Tribunal will then decide whether or not the trade mark should be cancelled in the UK.”

Given the ongoing value of trademark rights to modern businesses and the fact that the UK serves as an important trade hub for a number of international businesses operating in the EU, the UK should issue clear guidance regarding the grounds upon which a third-party challenge to a trademark application or registration will be accepted in the UK. Otherwise, businesses will be forced to inefficiently adjust their practices based upon the case-by-case rulings of the UK Intellectual Property Office, generating unnecessary instability for trademark owners in an already tumultuous time. 

Autonomous Vehicles May Never Become “Self-Aware”, But That Doesn’t Mean They’re Not Coming for Your Job

By: Mason Hudon

“Skynet begins to learn at a geometric rate. It becomes self-aware at 2:14 AM, Eastern time, August 29th. In a panic, they try to pull the plug.” – Terminator 2: Judgment Day

The Insidious Issue: Worker Displacement

True driverless cars will inevitably begin to find their way onto American roadways within the next decade, and the touted benefits are palpable. From saving time to reducing traffic fatalities, companies developing these technologies, including Tesla, Waymo, Ford, and Nissan, have rested their arguments for the development of driverless cars on reveries of a futuristic tomorrow in which one can press a button and be whisked off to their destination without a second thought. Sounds nice, doesn’t it?

It’s important to note that personal autonomous vehicles are likely not the first experience that the average American will have with this technology. Companies like Peloton Technology (no, not the exercise bike) and Locomation are making sure of that. These corporations are developing and refining a revolutionary process called “platooning” whereby two or more semi-trucks move together in a tight line while being piloted autonomously, with only the truck at the very front of the platoon piloted by a person. Platooning is poised to upend the freight industry and may lead to increased efficiency and decreased costs for major players in the space. According to Caleb Weaver, Uber’s Director of Public Affairs for the West Coast, the freight industry will undoubtedly be the first to feel a true and deep impact from AV innovation.

With every bit of news about the perils of autonomous vehicle technology causing accidental deaths, allowing source-code to be hacked, or increasing traffic-congestion in the short-to-medium term, comes reassurance from the industry that software will get better, cybersecurity will become more robust, and in the long-run, a driverless vehicle world will benefit society. Most of the time, manufacturers of autonomous vehicles respond to criticism with great aplomb. But, crucially, the autonomous vehicle world struggles to find answers to a serious criticism of the industry as a whole: widespread worker displacement in the transportation industry.

As the AV industry pushes ever forward, it falls to American lawmakers to act on this issue soon. The consequences of waiting much longer might be dire.

Despite a recent dearth of truck drivers in the United States, approximately 3.5 million people continue to drive freight, making it one of the largest industries for employment in the nation. In contrast, Uber, which is America’s most popular rideshare company, is said to contract with upwards of 1 million gig workers per year. Additionally there are about 207,000 taxi drivers in the United States. All told, this non-exhaustive list of drivers in the United States includes about 4.7 million people, many of whom may not have higher education or experience outside of their work as drivers. If these people lose their jobs to autonomous vehicles, lawmakers are going to be faced with a very tricky unemployment problem.

Potential Solutions

It’s doubtful that American lawmakers or industries will act as drastically as Indian Transport Minister Nitin Gadkari, who in 2017 vowed to prohibit driverless cars on Indian roads. Still, slowing the expansion of AV technologies may be part of the approach. In 2017, Microsoft magnate Bill Gates proposed a tax on robots that take jobs away from humans, and a similar measure was proposed in the European Union around the same time, although it ultimately failed to pass as law. In practice, a “robot tax” would have the effect of incentivizing companies to retain human workers for longer periods of time, and its proceeds could be used to fund other parts of the transition process, notably job retraining or universal basic income.

Job retraining is a particularly complex issue because it involves many different questions concerning different entities at every level of development and potential rollout. For example, should job retraining programs be funded by government entities or by the private corporations that are causing the worker displacement in the first place? Should these programs be formulated to give displaced workers total freedom of choice in deciding their “new careers” or should the programs be designed to focus on workers who intend to stay in the same industry? The latter question is especially poignant considering that “[w]hile new jobs will result from the new industry, they’re unlikely to be a direct match for the commercial driving ones that are going away. Those engineering and managing the technology, for example, are not the same folks that are driving buses.” In all likelihood, job retraining for potentially millions of displaced workers will require both private and public investment on a largely unprecedented scale, and the legal framework that will facilitate this process will have to be complex and developed in a timely manner to prevent an economic crisis for the displaced.

Universal basic income (UBI), like job retraining, presents unique challenges, but it places significant economic burdens on governmental entities rather than on private corporations, or worse yet, on private individuals. Indeed, UBI could significantly alleviate some of the growing pains associated with widespread worker displacement by autonomous vehicles. Federally, such a program is unlikely to ever come into existence. State governments, however, might be able to adopt a UBI system, provided they can figure out a viable way to fund it.

The No Solution Contingency

If the tech companies developing these technologies are to be believed, a solution (outside of simple job training) might not be necessary. According to Waymo, the elimination of jobs for drivers signals the creation of jobs for technicians, dispatchers, customer service representatives, and fleet response teams that will employ about as many people as are currently employed. These roles, which will likely become available at the majority of companies transitioning to autonomous vehicle fleets, can then be filled by people that are already company employees.

It sure sounds fantastic, but skepticism should remain high. Hoping that this situation will just work itself out (like Waymo suggests it will) is not a safe road for American lawmakers to take, and they should remain wary of the complexity of the issues at hand. Waymo’s approach downplays the costs and pitfalls of widespread job retraining, and it seems that many other autonomous vehicle companies are confident this issue will not be their cross to bear. Before we see actual job loss due to the AV industry, American lawmakers should begin preparing for the inevitable through proactive legislation and direct address of the issues.