Social Media

Whitelist for Thee, but Not for Me: Facebook File Scandals and Section 230 Solutions

Warren Sexson, MJLST Staffer

When I was in 7th grade, I convinced my parents to let me get my first social media account. Back in the stone age, that phrase was synonymous with Facebook. I never thought too much of how growing up in the digital age affected me, but looking back, it is easy to see the cultural red flags. It came as no surprise to me when, this fall, the Wall Street Journal broke what has been dubbed “The Facebook Files,” and in them found an internal study from the company showing Instagram is toxic to teen girls. While tragic, this conclusion is something many Gen-Zers and late-Millennials have known for years. However, in the “Facebook Files” there is another, perhaps even more jarring, finding: Facebook exempts many celebrities and elite influencers from its rules of conduct. This revelation demands a discussion of the legal troubles the company may find itself in and the proposed solutions to the “whitelisting” problem.

The Wall Street Journal’s reporting describes an internal process by Facebook called “whitelisting” in which the company “exempted high-profile users from some or all of its rules, according to company documents . . . .” This includes individuals from a wide range of industries and political viewpoints, from Soccer mega star Neymar, to Elizabeth Warren, and Donald Trump (prior to January 6th). The practice put the tech giant in legal jeopardy after a whistleblower, later identified as Frances Haugen, submitted a whistleblower complaint with the Securities and Exchange Commission (SEC) that Facebook has “violated U.S. securities laws by making material misrepresentations and omissions in statements to investors and prospective investors . . . .” See 17 CFR § 240.14a-9 (enforcement provision on false or misleading statements to investors). Mark Zuckerberg himself has made statements regarding Facebook’s neutral application of standards that are at direct odds with the Facebook Files. Regardless of the potential SEC investigation, the whitelist has opened up the conversation regarding the need for serious reform in the big tech arena to make sure no company can make lists of privileged users again. All of the potential solutions deal with 47 U.S.C. § 230, known colloquially as “section 230.”

Section 230 allows big tech companies to censor content while still being treated as a platform instead of a publisher (where they would incur liability for what is on their website). Specifically, § 230(c)(2)(A) provides that no “interactive computer service” shall be held liable for taking action in good faith to restrict “obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable [content] . . . .” It is the last phrase, “otherwise objectionable,” that tech companies have used as justification for removing “hate speech” or “misinformation” from their platform without incurring publisher like liability. The desire to police such speech has led Facebook to develop stringent platform rules which has in turn created the need for whitelisting. This brings us to our first proposal, eliminating the phrase “otherwise objectionable” from section 230 itself. The proposed “Stop the Censorship Act of 2020” brought by Republican Paul Gosar of Arizona does just that. Proponents argue that it would force tech companies to be neutral or lose liability protections. Thus, no big tech company would ever create standards stringent enough to require a “whitelist” or an exempted class, because the standard is near to First Amendment protections—problem solved! However, the current governing majority has serious concerns about forced neutrality, which would ignore problems of misinformation or the mental health effects of social media in the aftermath of January 6th.

Elizabeth Warren, similar to a recent proposal in the House Judiciary Committee, takes a different approach: breaking up big tech. Warren proposes passing legislation to limit big tech companies in competing with small businesses who use the platform and reversing/blocking mergers, such as Facebook purchasing Instagram. Her plan doesn’t necessarily stop companies from having whitelists, but it does limit the power held by Facebook and others which could in turn, make them think twice before unevenly applying the rules. Furthermore, Warren has called for regulators to use “every tool in the toolbox,” in regard to Facebook.

Third, some have claimed that Google, Facebook, and Twitter have crossed the line under existing legal doctrines to become state actors. So, the argument goes, government cannot “induce” or “encourage” private persons to do what the government cannot. See Norwood v. Harrison, 413 U.S. 455, 465 (1973). Since some in Congress have warned big tech executives to restrict what they see as bad content, the government has essentially co-opted the hand of industry to block out constitutionally protected speech. See Railway Employee’s Department v. Hanson, 351 U.S. 225 (1956) (finding state action despite no actual mandate by the government for action). If the Supreme Court were to adopt this reasoning, Facebook may be forced to adopt a First Amendment centric approach since the current hate speech and misinformation rules would be state action; whitelists would no longer be needed since companies would be blocked from policing fringe content. Finally, the perfect solution! The Court can act where Congress cannot agree. I am skeptical of this approach—needless to say, such a monumental decision would completely shift the nature of social media. While Justice Thomas has hinted at his openness to this argument, it is unclear if the other justices will follow suit.

All in all, Congress and the Court have tools at their disposal to combat the disturbing actions taken by Facebook. Outside of potential SEC violations, Section 230 is a complicated but necessary issue Congress must confront in the coming months. “The Facebook Files” have exposed the need for systemic change in social media. What I once used to use to play Farmville, has become a machine that has rules for me, but not for thee.


Lawyers in Flame Wars: The ABA Says Be Nice Online

Parker von Sternberg, MJLST Staffer

The advent of Web 2.0 around the turn of the millennium brought with it an absolute tidal wave of new social interactions. To this day we are in the throes of figuring out how to best engage with one another online, particularly when things get heated or otherwise out of hand. In this new wild west, lawyers sit at a perhaps unfortunate junction. Lawyers are indelibly linked to problems and moments of breakdown—precisely the events that lead to lashing out online. At the same time a lawyer, more so than many professions, relies upon their personal reputation to drive their business. When these factors collide, it creates pressure on the lawyer to defend themselves, but doing so properly can be a tricky thing.

When it comes to questions of ethics for lawyers, the first step is generally to crack open the Model Rules of Professional Conduct (MRPC), given that they have been adopted in 49 states and are kept up to date by the American Bar Association (ABA). While these model rules are customized to some extent from state to state, by and large the language used in the MRPC is an effective starting point for professional ethics issues across the country. Recently, the ABA has stepped into the fray with Formal Opinion 496, which lays out the official interpretation of MRPC 1.6 and how it comes into play in these situations.

MRPC 1.6 protects confidentiality of client information. For our purposes, the pertinent sections are

(a) A lawyer shall not reveal information relating to the representation of a client unless the client gives informed consent, the disclosure is impliedly authorized in order to carry out the representation or the disclosure is permitted by paragraph (b) and . . .

(b)(5) to establish a claim or defense on behalf of the lawyer in a controversy between the lawyer and the client, to establish a defense to a criminal charge or civil claim against the lawyer based upon conduct in which the client was involved, or to respond to allegations in any proceeding concerning the lawyer’s representation of the client.

So, when someone goes on Google Maps and excoriates your practice, a review that will pop up to everyone who even looks for directions to the office, what can be done? The first question is whether or not they are in fact a former client. If not, feel free to engage! Just wade in there and start publicly fighting with a stranger (really though, don’t do this. Even the ABA knows what the Streisand Effect is). However, if they are a former client, MRPC 1.6 or the local equivalent applies.

In Minnesota we have the MNRPC, with 1.6(b)(8) mirroring MRPC 1.6(b)(5). At its core, the ABA’s interpretation turns on the fact that an online review, on its own, does not qualify as a “controversy” or “proceeding.” That is not to say that it cannot give rise to one though! In 2016 an attorney in Florida took home $350,000 after former clients repeatedly defamed her because their divorce didn’t go how they wanted. But short of the outright lies in that case, lawyers suffering from online hit pieces are more limited in their options. The ABA lays out four possible responses to poor online reviews:

1) do not respond to the negative post or review at all, because as was brought up above, you tempt the Streisand Effect at your peril;

2) request that the website or search engine remove the review;

3) post an invitation to contact the lawyer privately to resolve the matter; and

4) indicate that professional considerations preclude a response.

While none of these options exactly inspire images of righteous fury, defending your besmirched professional honor or righting the wrongs done to your name, it appears unlikely that they will get you in trouble with the ethics board either. The ABA’s formal opinion lays out an impressive list of authorities from nearly a dozen states establishing that lawyers can and will face consequences for public review-related dust ups. The only option for an attorney really looking to have it out online it seems is to move to Washington D.C., where the local rules allow for disclosure “to the extent reasonably necessary to respond to specific allegations by the client concerning the lawyer’s representation of the client.”


Watching an APA Case Gestate Live!

Parker von Sternberg, MJLST Staffer

On October 15th the FCC published an official Statement of Chairman Pai on Section 230. Few particular statutes have come under greater fire in recent memory than the Protection for “Good Samaritan” Blocking and Screening of Offensive Material and the FCC’s decision to wade into the fray is almost certain to end up causing someone to bring suit regardless of which side of the issue the Commission comes down on.

As a brief introduction, 47 U.S. Code § 230 provides protections from civil suits for providers of Interactive Computer Services, which for our purposes can simply be considered websites. The statute was drafted and passed as a direct response by Congress to a pair of cases, namely Cubby, Inc. v. CompuServe Inc. and Stratton Oakmont, Inc. v. Prodigy Services Co.Cubby, Inc. v. CompuServe Inc., 776 F.Supp. 135 (S.D.N.Y. 1991) and Stratton Oakmont, Inc. v. Prodigy Services Co., 1995 WL 323710 (N.Y. Sup. Ct. 1995). Cubby held that the defendant, CompuServe, was not responsible for third-party posted content on its message board. The decisive reasoning by the court was that CompuServe was a distributor, not a publisher, and thus “must have knowledge of the contents of a publication before liability can be imposed.”Cubby, Inc. v. CompuServe Inc., 776 F.Supp. 135, 139 (S.D.N.Y. 1991). On the other hand, in Stratton Oakmont, the defendant’s exertion of “editorial control” over a message board otherwise identical to the one in Cubby “opened [them] up to a greater liability than CompuServe and other computer networks that make no such choice.” Stratton Oakmont, 1995 WL 323710 at *5.

Congress thus faced an issue: active moderation of online content, which is generally going to be a good idea, created civil liability where leaving message boards open as a completely lawless zone protects the owner of the board. The answer to this conundrum was § 230 which states, in part:

(c) Protection for “Good Samaritan” blocking and screening of offensive material

(1) Treatment of publisher or speaker

No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.

(2) Civil liability – No provider or user of an interactive computer service shall be held liable on account of—

(A) any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable, whether or not such material is constitutionally protected . . . .

Judicial application of the statute has so far largely read the language expansively. Zeran v. AOL held that “[b]y its plain language, § 230 creates a federal immunity to any cause of action that would make service providers liable for information originating with a third-party user of the service.”Zeran v. Am. Online, Inc., 129 F.3d 327, 330 (4th Cir. 1997). The court also declined to recognize a difference between a defendant acting as a publisher versus a distributor. Speaking to Congress’s legislative intent, the court charted a course that aimed to both immunize service providers as well as encourage self-regulation. Id. at 331-334. Zeran has proved immensely influential, having been cited over a hundred times in the ensuing thirteen years.

Today however, the functioning of § 230 has become a lightning rod for the complaints of many on social media. Rather than encouraging interactive computer services to self-regulate, the story goes that it instead protects them despite their “engaging in selective censorship that is harming our national discourse.” Republicans in the Senate have introduced a bill to amend the Communications Decency Act specifically to reestablish liability for website owners in a variety of ways that § 230 currently protects them from. The Supreme Court has also dipped its toes in the turbulent waters of online censorship fights, with Justice Thomas saying that “courts have relied on policy and purpose arguments to grant sweeping protection to Internet platforms” and that “[p]aring back the sweeping immunity courts have read into §230 would not necessarily render defendants liable for online misconduct.

On the other hand, numerous private entities and individuals hold that § 230 forms part of the backbone of the internet as we know it today. Congress and the courts, up until a couple of years ago, stood in agreement that it was vitally important to draw a bright line between the provider of an online service and those that used it. It goes without saying that some of the largest tech companies in the world directly benefit from the protections offered by this law, and it can be argued that the economic impact is not limited to those larger players alone.

What all of this hopefully goes to show is that, no matter what happens to this statute, someone somewhere will be willing to spend the time and the money necessary to litigate over it. The question is what shape that litigation will take. As it currently stands, the new bill in the Senate has little chance of getting through the House of Representatives to the President’s desk. The Supreme Court just recently denied cert to yet another § 230 case, upholding existing precedent. Enter Ajit Pai and the FCC, with their legal authority to interpret 47 U.S. Code § 230. Under the cover of Chevron deference protecting administrative action with regard to interpreting statutes the legislature has empowered them to enforce, the FCC wields massive influence with regard to the meaning of § 230. Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984).

While the FCC’s engagement is currently limited to a statement that it intends to “move forward with rulemaking to clarify [§ 230’s] meaning,” there are points to discuss. What limits are there on the power to alter the statute’s meaning? Based on the Commissioner’s statement, can we tell generally what side they are going to come down on? With regard to the former, as was said above, the limit is set largely by Chevron deference and by § 706 of the APA. The key words here are going to be if whoever ends up unhappy with the FCC’s interpretation can prove that it is “arbitrary and capricious” or goes beyond a “permissible construction of the statute.” Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984).

The FCC Chairman’s statement lays out that issues exist surrounding §230 and establishes that the FCC believes the legal authority exists for it to interpret the statute. It finishes by saying “[s]ocial media companies have a First Amendment right to free speech. But they do not have a First Amendment right to a special immunity denied to other media outlets, such as newspapers and broadcasters.” Based on this statement alone, it certainly sounds like the FCC intends to narrow the protections for interactive computer services providers in some fashion. At the same time, it raises questions. For example, does § 230 provide websites with special forms of free speech that other individuals and groups do not have? The statute does not on its face make anything legal that without it would not be. Rather, it ensures that legal responsibility for speech lies with the speaker, rather than the digital venue in which it is said.

The current divide on liability for speech and content moderation on the internet draws our attention to issues of power as the internet continues to pervade all aspects of life. When the President of the United States is being publicly fact-checked, people will sit up and take notice. The current Administration, parts of the Supreme Court, some Senators, and now the FCC all appear to feel that legal proceedings are a necessary response to this happening. At the same time, alternative views do exist outside of Washington D.C., and at many points they may be more democratic than those proposed within our own government.

There is a chance that if the FCC takes too long to produce a “clarification” of §230 that Chairman Pai will be replaced after the upcoming Presidential election. Even if this does happen, I feel that the outlining of the basic positions surrounding this statute is nonetheless worthwhile. A change in administrations simply means that the fight will occur via proposed statutory amendments or in the Supreme Court, rather than via the FCC.

 


Privacy, Public Facebook Posts, and the Medicalization of Everything

Peter J. Teravskis, MD/JD Candidate, MJLST Staffer

Medicalization is “a process by which human problems come to be defined and treated as medical problems.” Medicalization is not a formalized process, but is instead “a social meaning embedded within other social meanings.” As the medical domain has expanded in recent years scholars have begun to point to problems with “over-medicalization” or “corrupted medicalization.” Specifically, medicalization is used to describe “the expansion of medicine in people’s lives.” For example, scholars have problematized the medicalization of obesity, shynesshousing, poverty, normal aging, and even dying, amongst many others. The process of medicalization has become so pervasive in recent years that various sociologists have begun to discuss it as the medicalization “of everyday life,” “of society,”  “of culture,” of the human condition, and “the medicalization of everything”—i.e. turning all human difference into pathology. Similarly, developments in “technoscientific biomedicine” have led scholars to blur the line of what is exclusively “medical” into a broader process of “biomedicalization.”

Medicalization does not carry a valence of “good” or “bad” per se: medicalization and demedicalization can both restrict and expand personal liberties. However, when everyday living is medicalized there are many attendant problems. First, medicalization places problems outside a person’s control: rather than the result of choice, personality, or character, a medicalized problem is considered biologically preordained or “curable.” Medicalized human differences are no longer considered normal; therefore, “treatment” becomes a “foregone conclusion.” Because of this, companies are incentivized to create pharmacological and biotechnological solutions to “cure” the medicalized problem. From a legal perspective, Professor Adele E. Clarke and colleagues note that through medicalization, “social problems deemed morally problematic . . . [are] moved from the professional jurisdiction of the law to that of medicine.” This process is referred to, generally, as the “medicalization of deviance.” Further, medicalization can de-normalize aspects of the human condition and classify people as “diseased.”

Medicalization is important to the sociological study of social control. Social control is defined as the “mechanisms, in the form of patterns of pressure, through which society maintains social order and cohesion.” Thus, once medicalized, an illness is subject to control by medicinal interventions (drugs, surgery, therapy, etc.) and a sick people are expected to take on the “sick role” whereby they become the subjects of physicians’ professional control. A recent example of medical social control is the social pressure to engage in hygienic habits, precautionary measures, and “social distancing” in response to the novel coronavirus, COVID-19. The COVID-19 pandemic is an expressly medical problem; however, when normal life, rather than a viral outbreak, is medicalized, medical social control becomes problematic. For example, the sociologist Peter Conrad argues that medical social control can take the form of “medical surveillance.” He states that “this form of medical social control suggests that certain conditions or behaviors become perceived through a ‘medical gaze’ and that physicians may legitimately lay claim to all activities concerning the condition” (quoting Michel Foucault’s seminal book The Birth of the Clinic).

The effects of medical social control are amplified due to the communal nature of medicine and healthcare, leading to “medical­legal hybrid[]” social control and, I argue, medical-corporate social control. For example, employers and insurers have interests in encouraging healthful behavior when it reduces members’ health care costs. Similarly, employers are interested in maximizing healthy working days, decreasing worker turnover, and maximizing healthy years, thus expanding the workforce. The State has similar interests, as well as interests in reducing end-of-life and old age medical costs. At first glance, this would seem to militate against overmedicalization. However, modern epidemiological methods have revealed the long term consequences of untreated medical problems. Thus, medicalization may result in the diversion of health care dollars towards less expensive preventative interventions and away from more expensive therapy that would help later in life.

An illustrative example is the medicalization of obesity. Historically, obesity was not considered a disease but was a socially desirable condition: demonstrating wealth; the ability to afford expensive, energy-dense foods; and a life of leisure rather than manual labor. Changing social norms, increased life expectancy, highly sensitive biomedical technologies for identifying subtle metabolic changes in blood chemistry, and population-level associations between obesity and later-life health complications have contributed to the medicalization of this conditions. Obesity, unlike many other conditions, it not attributable to a single biological process, rather, it is hypothesized to result from the contribution of multiple genetic and environmental factors. As such, there is no “silver bullet” treatment for obesity. Instead, “treatment” for obesity requires profound changes reaching deep into how a patient lives her life. Many of these interventions have profound psychosocial implications. Medicalized obesity has led, in part, to the stigmatization of people with obesity. Further, medical recommendations for the treatment of obesity, including gym membership, and expensive “health” foods, are costly for the individual.

Because medicalized problems are considered social problems affecting whole communities, governments and employers have stepped in to treat the problem. Politically, the so-called “obesity epidemic” has led to myriad policy changes and proposals. Restrictions designed to combat the obesity epidemic have included taxes, bans, and advertising restrictions on energy-dense food products. On the other hand, states and the federal government have implemented proactive measures to address obesity, for example public funds have been allocated to encourage access to and awareness of “healthy foods,” and healthy habits. Further, Social Security Disability, Medicare and Medicaid, and the Supplemental Nutrition Assistance Program have been modified to cope with economic and health effects of obesity.

Other tools of control are available to employers and insurance providers. Most punitively, corporate insurance plans can increase rates for obese employees.  As Abby Ellin, writing for Observer, explained “[p]enalizing employees for pounds is perfectly legal [under the Affordable Care Act]” (citing a policy brief published in the HealthAffairs journal). Alternatively, employers and insurers have paid for or provided incentives for gym memberships and use, some going so far as to provide exercise facilities in the workplace. Similarly, some employers have sought to modify employee food choices by providing or restricting food options available in the office. The development of wearable computer technologies has presented another option for enforcing obesity-focused behavioral control. Employer-provided FitBits are “an increasingly valuable source of workforce health intelligence for employers and insurance companies.” In fact, Apple advertises Apple Watch to corporate wellness divisions and various media outlets have noted how Apple Watch and iPhone applications can be used by employers for health surveillance.

Indeed, medicalization as a pretense for technological surveillance and social control is not exclusively used in the context of obesity prevention. For instance, the medicalization of old age has coincided with the technological surveillance of older people. Most troubling, medicalization in concert with other social forces have spawned an emerging field of technological surveillance of mental illness. Multiple studies, and current NIH-funded research, are aimed at developing algorithms for the diagnosis of mental illness based on data mined from publicly accessible social media and internet forum posts. This process is called “social media analysis.” These technologies are actively medicalizing the content of digital communications. They subject peoples’ social media postings to an algorithmic imitation of the medical gaze, whereby, “physicians may legitimately lay claim to” those social media interactions.  If social media analysis performs as hypothesized, certain combinations of words and phrases will constitute evidence of disease. Similar technology has already been coopted as a mechanism of social control to detect potential perpetrators of mass shootings. Policy makers have already seized upon the promise of medical social media analysis as a means to enforce “red flag” laws. Red flag laws “authorize courts to issue a special type of protection order, allowing the police to temporarily confiscate firearms from people who are deemed by a judge to be a danger to themselves or to others.” Similarly, it is conceivable that this type of evidence will be used in civil commitment proceedings. If implemented, such programs would constitute a link by which medical surveillance, under the banner of medicalization, could be used as grounds to deprive individuals of civil liberty, demonstrating an explicit medical-legal hybrid social control mechanism.

What protections does the law offer? The Fourth Amendment protects people from unreasonable searches. To determine whether a “search” has occurred courts ask whether the individual has a “reasonable expectation of privacy” in the contents of the search. Therefore, whether a person had a reasonable expectation of privacy in publicly available social media data is critical to determining whether that data can be used in civil commitment proceedings or for red flag law protective orders.

Public social media data is, obviously, public, so courts have generally held that individuals have no reasonable expectation of privacy in its contents. By contrast, the Supreme Court has ruled that individuals have a reasonable expectation of privacy in the data contained on their cell phones and personal computers, as well as their personal location data (cell-site location information) legally collected by third party cell service providers. Therefore, it is an open question how far a person’s reasonable expectation of privacy extends in the case of digital information. Specifically, when public social media data is used for medical surveillance and making psychological diagnoses the legal calculation may change. One interpretation of the “reasonable expectation of privacy” test argues that it is an objective test—asking whether a reasonable person would actually have a privacy interest. Indeed, some scholars have suggested using polling data to define the perimeter of Fourth Amendment protections. In that vein, an analysis of the American Psychiatric Association’s “Goldwater Rule” is illustrative.

The Goldwater Rule emerged after the media outlet “Fact” published psychiatrists’ medical impressions of 1964 presidential candidate Barry Goldwater. Goldwater filed a libel suit against Fact, and the jury awarded him $1.00 in compensatory damages and $75,000 in punitive damages resulting from the publication of the psychiatric evaluations. None of the quoted psychiatrists had met or examined Goldwater in person. Subsequently, concerned primarily about the inaccuracies of “diagnoses at a distance,” the APA adopted the Goldwater Rule, prohibiting psychiatrists from engaging in such practices. It is still in effect today.

The Goldwater Rule does not speak to privacy per se, but it does speak to the importance of personal, medical relationships between psychiatrists and patients when arriving at a diagnosis. Courts generally treat those types of relationships as private and protect them from needless public exposure. Further, using social media surveillance to diagnose mental illness is precisely the type of diagnosis-at-a-distance that concerns the APA. However, big-data techniques promise to obviate the diagnostic inaccuracies the 1960s APA was concerned with.

The jury verdict in favor of Goldwater is more instructive. While the jury found only nominal compensatory damages, it nevertheless chose to punish Fact magazine. This suggests that the jury took great umbrage with the publication of psychiatric diagnoses, even though they were obtained from publicly available data. Could this be because psychiatric diagnoses are private? The Second Circuit, upholding the jury verdict, noted that running roughshod over privacy interests is indicative of malice in cases of libel. Under an objective test, this seems to suggest that subjecting public information to the medical gaze, especially the psychiatrist’s gaze, unveils information that is private. In essence, applying big-data computer science techniques to public posts unveils or reveals private information contained in the publicly available words themselves. Even though the public social media posts are not subject to a reasonable expectation of privacy, a psychiatric diagnosis based on those words may be objectively private. In sum, the medicalization and medical surveillance of normal interactions on social media may create a Fourth Amendment privacy interest where none previously existed.


Zoinks! Can the FTC Unmask Advertisements Disguised by Social Media Influencers?

Jennifer Satterfield, MJLST Staffer

Social media sites like Instagram and YouTube are filled with people known as “influencers.” Influencers are people with a following on social media that use their online fame to promote products and services of a brand. But, with all that power comes great responsibility, and influencers, as a whole, are not being responsible. One huge example of irresponsible influencer activity is the epic failure and fraudulent music festival known as Fyre Festival. Although Fyre Festival promised a luxury, VIP experience on a remote Bahamian island, it was a true nightmare where “attendees were stranded with half-built huts to sleep in and cold cheese sandwiches to eat.” The most prominent legal action was against Fyre’s founders and organizers, Billy McFarland and Ja Rule, including a six-year criminal sentence for wire fraud against McFarland. Nonetheless, a class action lawsuit also targeted the influencers. According to the lawsuit, the influencers did not comply with Federal Trade Commission (“FTC”) guidelines and disclose they were being paid to advertise the festival. Instead, “influencers gave the impression that the guest list was full of the Social Elite and other celebrities.” Yet, the blowback against influencers since the Fyre Festival fiasco appears to be minimal.

According to a Mediakix report, “[i]n one year, a top celebrity will post an average of 58 sponsored posts and only 3 may be FTC compliant.” The endorsement guidelines specify that if there is a “material connection” between the influencer and the seller of an advertised product, this connection must be fully disclosed. The FTC even created a nifty guide for influencers to ensure compliance. While disclosure is a small burden and there are several resources informing influencers of their duty to disclose, these guidelines are still largely ignored.

Evens so, the FTC has sent several warning letters to individual influencers over the years, which indicates it is monitoring top influencers’ posts. However, a mere letter is not doing much to stop the ongoing, flippant, and ignorant disregard toward the FTC guidelines. Besides the letters, the FTC rarely takes action against individual influencers. Instead, if the FTC goes after a bad actor, “it’s usually a brand that[] [has] failed to issue firm disclosure guidelines to paid influencers.” Consequently, even though it appears as if the FTC is cracking down on influencers, it is really only going after the companies. Without actual penalties, it is no wonder most influencers are either unaware of the FTC guidelines or continue to blatantly ignore them.

Considering this problem, there is a question of what the FTC can really do about it. One solution is for the FTC to dig in and actually enforce its guidelines against influencers like it did in 2017 with CSGO Lotto and two individual influencers, Trevor Martin and Thomas Cassell. CSGO Lotto was a website in which users could gamble virtual items called “skins” from the game Counter-Strike: Global Offensive. According to the FTC’s complaint, Martin and Thomas endorsed CSGO Lotto but failed to disclose they were both the owners and officers of the company. CSGO Lotto also paid other influencers to promote the website. The complaint notes that numerous YouTube videos by these influencers either failed to include a sponsorship disclosure in the videos or inconspicuously placed such disclosures “below the fold” in the description box. While the CSGO Lotto action was a huge scandal in the video game industry, it was not widely publicized to the general population. Moreover, Martin and Cassell got away with a mere slap on the wrist—“[t]he [FTC] order settling the charges requires Martin and Cassell to clearly and conspicuously disclose any material connections with an endorser or between an endorser and any promoted product or service.” Thus, it was not enough to compel other influencers into compliance. Instead, if the FTC started enforcement actions against big-name influencers, other influencers may also fear retribution and comply.

On the other hand, the FTC could continue its enforcement against the companies themselves, but this time with more teeth. Currently, the FTC is preparing to take further steps to ensure consumer protection in the world of social media influencers. Recently, FTC Commissioner Rohit Chopra acknowledged in a public statement that “it is not clear whether our actions are deterring misconduct in the marketplace, due to the limited sanctions we have pursued.” Although Chopra is not interested in pursuing small influencers, but rather the advertisers that pay them, it is possible that enforcement against the companies will cause influencers to comply as well.

Accordingly, Chopra’s next steps include: (1) “[d]eveloping requirements for technology platforms (e.g. Instagram, YouTube, and TikTok) that facilitate and either directly or indirectly profit from influencer marketing;” (2) “[c]odifying elements of the existing endorsement guides into formal rules so that violators can be liable for civil penalties under Section 5(m)(1)(A) and liable for damages under Section 19; 7;” and (3) “[s]pecifying the requirements that companies must adhere to in their contractual arrangements with influencers, including through sample terms that companies can include in contracts.” By pushing some of the enforcement duties onto social media platforms themselves, the FTC gains more monitoring and enforcement capabilities. Furthermore, codifying the guidelines into formal rules gives the FTC teeth to impose civil penalties and creates tangible consequences for those who previously ignored the guidelines. Finally, by actually requiring companies to adhere to these rules via their contract with influencers, influencers will be compelled to follow the guidelines as well. Therefore, under these next steps, paid advertising disclosures on social media can become commonplace. But only time will really tell if the FTC will achieve these steps.


Google Fined for GDPR Non-Compliance, Consumers May Not Like the Price

Julia Lisi, MJLST Staffer

On January 14th, 2019, France’s Data Protection Authority (“DPA”) fined Google 50 million euros in one of the first enforcement actions taken under the EU’s General Data Protection Regulation (“GDPR”). The GDPR, which took effect in May of 2018, sent many U.S. companies scrambling in attempts to update their privacy policies. You, as a consumer, probably had to re-accept updated privacy policies from your social media accounts, phones, and many other data-based products. Google’s fine makes it the first U.S. tech giant to face GDPR enforcement. While a 50 million euro (roughly 57 million dollars) fine may sound hefty, it is actually relatively small compared to maximum fine allowed under the GDPR, which, for Google, would be roughly five billion dollars.

The French fine clarifies a small portion of the uncertainty surrounding GDPR enforcement. In particular, the French DPA rejected Google’s methods for getting consumers to consent to its  Privacy Policy and Terms of Service. The French DPA took issue with the (1) numerous steps users faced before they could opt out of Google’s data collection, (2) the pre-checked box indicating users’ consent, and (3) the inability of users to consent to individual data processes, instead requiring whole cloth acceptance of both Google’s Privacy Policy and Terms of Service.

The three practices rejected by the French DPA are commonplace in the lives of many consumers. Imagine turning on your new phone for the first time and scrolling through seemingly endless provisions detailing exactly how your daily phone use is tracked and processed by both the phone manufacturer and your cell provider. Imagine if you had to then scroll through the same thing for each major app on your phone. You would have much more control over your digital footprint, but would you spend hours reading each provision of the numerous privacy policies?

Google’s fine could mark the beginning of sweeping changes to the data privacy landscape. What once took a matter of seconds—e.g., checking one box consenting to Terms of Service—could now take hours. If Google’s fine sets a precedent, consumers could face another wave of re-consenting to data use policies, as other companies fall in line with the GDPR’s standards. While data privacy advocates may applaud the fine as the dawn of a new day, it is unclear how the average consumer will react when faced with an in-depth consent process.


Tinder Shows Discrimination Can Take All Shapes In The Internet Age

Caleb Holtz, MJLST Staffer

On January 20th Tinder Inc., the company responsible for the popular dating mobile app, filed a proposed settlement agreement worth over $17 million. The settlement seeks to settle claims that Tinder charged older users more to use the app solely because of their age. Interestingly, while many people think of age discrimination against a group for being too old as being solely the concern of AARP members, this discrimination was against people over the age of 29. This is because of the relatively low threshold in California as to what can constitute age discrimination under California civil rights and consumer protection laws.

Discrimination is incredibly common in the Internet age, at least partially because it is so easy to do. Internet users develop a digital “fingerprint” over time and usage which follows them from website to website. Data contained within a digital “fingerprint” can contain information from “websites you visit, social platforms you use, searches you perform, and content you consume.” Digital fingerprinting is becoming even more common, as enterprising trackers have discovered a way to track users across multiple different browsing applications. When this information is combined with data users willfully give out on the internet, such as personal data on Facebook or Tinder, it is incredibly easy for companies to create a profile of all of the users relevant characteristics. From there it is easy to choose on what grounds to distinguish, or discriminate, users.

Discrimination in this manner is not always necessarily bad. On the most positive end of the spectrum, institutions like banks can use the information to discern if the wrong person is trying to access an account, based on the person’s digital fingerprint. More commonly, internet companies use the data to discriminate against users, controlling what they see and the price they are offered. A quintessential example of this practice was the study that found travel websites show higher prices to Mac users than PC users. Advocates of the practice argue that it allows companies to customize the user experience on an individual basis, allowing the user to see only what they want to see. They also say that it allows businesses to maximize efficiency, both in terms of maximizing profits and in terms of catering to the customer flow, which would therefore lead to a better user experience in the long run. To this point, the argument in favor of continuing this practice has generally won out, as it remains generally legal in the United States.

Opponents of the practice however say the costs outweigh the benefits. Many people, when shown just how much personal data follows them around the internet, will find the practice “creepy”. Opponents hope they can spread this general sentiment by showing more people just how much of their data is online without their explicit consent. This idea has support because, “despite its widespread use, price discrimination is largely happening without the knowledge of the general public, whose generally negative opinion of the practice has yet to be heard.”

More serious opponents move past the “creepiness” and into the legal and ethical issues that can pop up. As the Tinder case demonstrates, online discrimination can take an illegal form, violating state or federal law. Discrimination can also be much more malicious, allowing for companies looking for new employees to choose who even sees the job opening, based on factors like race, age, or gender. As Gillian B. White recently summarized nicely, “while it’s perfectly legal to advertise men’s clothing only to men, it’s completely illegal to advertise most jobs exclusively to that same group.” Now, as the Tinder case demonstrates, in certain scenarios it may be illegal to discriminate in pricing as well as job searches.

So what can be done about this, from a legal perspective? Currently in the United States the main price discrimination laws, the Sherman Antitrust Act, the Clayton Act, and the Robinson-Patman Act were created long before the advent of the internet, and allow for price discrimination as long as there is a “good faith reason”. (Part of the trouble Tinder ran into in litigation is a judge’s finding that there was not a good faith reason to discriminate as they were). There are also a plethora of discrimination in hiring laws which make certain discrimination by hiring employers illegal. Therefore the best current option may be for internet watchdog groups to keep a keen eye out for these practices and report what they come across.

As far as how the law can be changed, an interesting option exists elsewhere in the world. European Union data privacy laws may soon make some price discrimination illegal, or at the very least, significantly more transparent so users are aware of how their data is being used. Perhaps by similarly shining sunlight on the issue here in the states, consumers will begin forcing companies to change their practices.


Controversial Anti-Sex Trafficking Bill Eliminates Safe-Harbor for Tech Companies

Maya Digre, MJLST Staffer

 

Last week the U.S. Senate voted to approve the Stop Enabling Sex Traffickers Act. The U.S. House of Representatives also passed a similar bill earlier this year. The bill creates an exception to Section 230 of the Communications Decency Act that allows victims of sex trafficking to sue websites that enabled their abuse. The bill was overwhelmingly approved in both the U.S. House and Senate, receiving 388-25 and 97-2 votes respectively. President Trump has indicated that he is likely to sign the bill.

 

Section 230 of the Communications Decency Act shields websites from liability stemming from content posted by third parties on their sites. Many tech companies argue that this provision has allowed them to become successful without a constant threat of liability. However, websites like Facebook, Google, and Twitter have recently received criticism for the role they played in unintentionally meddling in the 2016 presidential election. Seemingly the “hands off” approach of many websites has become a problem that Congress now seeks to address, at least with respect to sex trafficking.

 

The proposed exception would expose websites to liability if they “knowingly” assist, support, or facilitate sex trafficking. The bill seeks to make websites more accountable for posts on their site, discouraging a “hands off” approach.

 

While the proposed legislation has received bipartisan support from congress, it has been quite controversial in many communities. Tech companies, free-speech advocates, and consensual sex workers all argue that the bill will have unintended adverse consequences. The tech companies and free-speech advocates argue that the bill will stifle speech on the internet, and force smaller tech companies out of business for fear of liability. Consensual sex workers argue that this bill will shut down their online presence, forcing them to engage in high-risk street work. Other debates center on how the “knowingly” standard will affect how websites are run. Critics argue that, in response to this standard, “[s]ites will either censor more content to lower risk of knowing about sex trafficking, or they will dial down moderation in an effort not to know.” At least one website has altered their behavior in the wake of this bill. In response to this legislation Craigslist has remove the “personal ad” platform from their website.

 


Judicial Interpretation of Emojis and Emoticons

Kirk Johnson, MJLST Staffer

 

In 2016, the original 176 emojis created by Shigetaka Kurita were enshrined in New York’s Museum of Modern Art as just that: art. Today, a smartphone contains approximately 2,000 icons that many use as a communication tool. New communicative tools present new problems for users and the courts alike; when the recipient of a message including an icon interprets the icon differently than the sender, how should a court view that icon? How does it affect the actus reus or mens rea of a crime? While a court has a myriad of tools that they use to decipher the meaning of new communicative tools, the lack of a universal understanding of these icons has created interesting social and legal consequences.

The first of many problems with the use of an emoji is that there is general disagreement on what the actual icon means. Take this emoji for example: 🙏. In a recent interview by the Wall Street Journal, people aged 10-87 were asked what this symbol meant. Responses varied from hands clapping to praying. The actual title of the emoji is “Person with Folded Hands.”

Secondly, the icons can change over time. Consider the update of the Apple iOS from 9 to 10; many complained that this emoji, 💁, lost its “sass.” It is unclear whether the emoji was intended to have “sass” to begin with, especially since the title of the icon is “Information Desk Person.”

Finally, actual icons vary from device to device. In some instances, when an Apple iPhone user sends a message to an Android phone user, the icon that appears on the recipient’s screen is completely different than what the sender intended. When Apple moved from iOS 9 to iOS 10, they significantly altered their pistol emoji. While an Android user would see something akin to this 🔫, an iPhone user sees a water pistol. Sometimes, an equivalent icon is not present on the recipient’s device and the only thing that appears on their screen is a black box.

Text messages and emails are extremely common pieces of evidence in a wide variety of cases, from sexual harassment litigation to contract disputes. Recently, the Ohio Court of Appeals was called upon to determine whether the text message “come over” with a “winky-face emoji” was adequate evidence to prove infidelity. State v. Shepherd, 81 N.E.3d 1011, 1020 (Ohio Ct. App. 2017). A Michigan sexual harassment attorney’s client was convinced that an emoji that looked like a horse followed by an icon resembling a muffin meant “stud muffin,” which the client interpreted as an unwelcome advance from a coworker. Luckily, messages consisting entirely of icons rarely determine the outcome of a case on their own; in the sexual harassment arena, a single advance from an emoji message would not be sufficient to make a case.

However, the implications are much more dangerous in the world of contracts. According to the Restatement (Second) of Contracts § 20 (1981),

(1) There is no manifestation of mutual assent to an exchange if the parties attach materially different meanings to their manifestations and

(a) neither party knows or has reason to know the meaning attached by the other; or

(b) each party knows or each party has reason to know the meaning attached by the other.

(2) The manifestations of the parties are operative in accordance with the meaning attached to them by one of the parties if

(a) that party does not know of any different meaning attached by the other, and the other knows the meaning attached by the first party; or

(b) that party has no reason to know of any different meaning attached by the other, and the other has reason to know the meaning attached by the first party.

 

Adhering to this standard with emojis would produce varied and unexpected results. For example, if Adam sent Bob a message “I’ll give you $5 to mow my lawn 😉,” would Bob be free to accept the offer? Would the answer be different if Adam used the 😘 emoji instead of the 😉 emoji? What if Bob received a black box instead of any emoji at all? Conversely, if Adam sent Bob the message without an emoji and Bob replied to Adam “Sure 😉,” should Adam be able to rely upon Bob’s message as acceptance? In 2014, the Michigan Court of Appeals ruled that the emoticon “:P” denoted sarcasm and that the text prior to the message should be interpreted with sarcasm. Does this extend to the emoji 😜😝, and 😛, titled “Face with Stuck-Out Tongue And Winking Eye,” “Face With Stuck-Out Tongue And Tightly-Closed Eyes,” and “Face With Stuck-Out Tongue” respectively?

In a recent case in Israel, a judge ruled that the message “✌👯💃🍾🐿☄constituted acceptance of a rental contract. While the United States does have differing standards for the laws of contracts, it seems that a judge could find that to be acceptance under the Restatement of Contracts (Second) § 20(2). Eric Goldman at the Santa Clara University School of Law hypothesizes that an emoji dictionary might help alleviate this issue. While a new Black’s Emoji Law Dictionary may seem unnecessary to many, without some sort of action it will be the courts deciding what the meaning of an emoji truly is. In a day where courts rule that a jury is entitled to actually see the emoji rather than have a description read to them, we can’t ignore the reality that action is necessary.


E-threat: Imminent Danger in the Information Age

MJLST Staffer, Jacob Weindling

 

One of the basic guarantees of the First Amendment is the right to free speech. This right protects the individual from restrictions on speech by the government, but is often invoked as a rhetorical weapon against private individuals or organizations declining to publish another’s words. On the internet, these organizations include some of the most popular discussion platforms in the U.S. including Facebook, Reddit, Yahoo, and Twitter. A key feature of these organizations is their lack of government control. As recenty as 2017, the Supreme Court has identified First Amendment grounds for overturning prohibitions on social media access. Indeed, one of the only major government prohibitions on speech currently in force is the ban on child pornography. Violent rhetoric, meanwhile, continues to fall under the Constitutional protections identified by the Court.

Historically, the Supreme Court has taken a nuanced view of violent speech as it relates to the First Amendment. The Court held in Brandenburg v. Ohio that “the constitutional guarantees of free speech and free press do not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.” Contrast this with discussion of a moral responsibility to resort to violence, which the Supreme Court has held to be distinct from preparing a group for imminent violent acts.

With the rise and maturation of the internet, public discourse has entered a new and relatively unchartered territory that the Supreme Court would have been hard-pressed to anticipate at the time of the Brandenburg and Noto decisions. Where once geography served to isolate Neo-Nazi groups and the Ku Klux Klan into small local chapters, the internet now provides a centralized meeting place for the dissemination and discussion of violent rhetoric. Historically, the Supreme Court concerned itself mightily with the distinction between an imminent call to action and a general discussion of moral imperatives, making clear delineations between the two.

The context of the Brandenburg decision was a pre-information age telecommunications regime. While large amounts of information could be transmitted around the world in relatively short order thanks to development of international commercial air travel, real-time communication was generally limited to telephone conversations between two individuals. An imminent call to action would require substantial real-world logistics, meetings, and preparation, all of which provide significant opportunities for detection and disruption by law enforcement. By comparison, internet forums today provide for near-instant communication between large groups of individuals across the entire world, likely narrowing the window that law enforcement would have to identify and act upon a credible, imminent threat.

At what point does Islamic State recruitment or militant Neo-Nazi organizing on the internet rise to the level of imminent threat? The Supreme Court has not yet decided the issue, many internet businesses have recently begun to take matters into their own hands. Facebook and Youtube have reportedly been more active in policing Islamic State propaganda, while Reddit has taken some steps to remove communities that advocate for rape and violence. Consequently, while the Supreme Court has not yet elected to draw (or redraw) a bright red line in the internet age, many businesses appear to be taking the first steps to draw the line themselves, on their terms.