Archive – 15 Sep 2023

UK Finally Producing Legislation to Establish CMA’s Digital Markets Unit

Report from the Financial Times

In Brief –  The UK Government has released a draft of the Digital Markets, Competition and Consumer Bill, legislation fully establishing the regulatory Digital Markets Unit (DMU). The DMU will be housed in the UK Competition and Markets Authority, the country’s competition regulator. The legislation gives the regulator authority to designate large digital businesses as having “strategic market status” (SMS) if they are adjudged to hold entrenched market power in at least one digital activity and have global sales above $31.2 billion (or UK revenue above £1 billion). SMS firms will be subject to a code of conduct enforced by the DMU, with fines of up to 10% of global turnover and authority to order changes. Merger guidelines are toughened for the digital giants, including lower notification thresholds and more stringent tests to assess the likelihood of deals causing future harm. Finally, the draft adds to consumer protection rules, including on data use, terms and conditions, fake online reviews, and the cancellation of subscription contracts

Context –  The Falcon said of Captain America, “I do what he does, only slower”. The UK CMA seems to be playing that role in relation to the EU, and even continental colleagues such as Germany’s Federal Cartel Office (FCO). In 2021, the German legislature responded to claims that the traditional competition enforcement model was far too slow to deal with digital giants by granting the FCO authority to proactively regulate the largest digital platforms. The regulator has ruled that Google, Meta, Amazon, and Apple fall under the new hybrid antitrust-regulatory authority. The EU enacted it’s Big Tech regulatory regime, the Digital Market Act (DMA), last year. It establishes a code of conduct with 18 high-level “Do’s and Don’ts” for “digital gatekeepers”. With its new DMU legislation, and the DSA-like Online Safety Bill, the UK clearly does not want to be left out of the regulatory party, although the desire to be a hub for technology investment and innovation seems increasingly unlikely. On the other hand, US progressives can only express frustration over not being able to join in directly.

Brazil Steps to the Forefront in the Social Media Content Moderation Debate

Report from the Brazilian Report

In Brief –  The Brazilian Congress is moving forward with special expedited consideration of legislation to hold digital platforms accountable for their content moderation decisions regarding harmful content, online radicalization, and misinformation. The Brazilian Government and other advocates cite a handful of recent school attacks as well as the violent riot and storming of the national capital facility in January following the election of Lula da Silva as President of Brazil. The Federal Supreme Court recently held a public hearing regarding the constitutionality of Article 19 of the Marco Civil da Internet, Brazil’s landmark internet civil rights framework, which establishes that digital platforms can only be held liable for content after receiving a court order instructing them to remove it, except for copyrighted material and non-consensual intimate images. At the hearing, the Government’s Special Advisor for Digital Rights argued that Article 19’s liability safe harbor was created under the assumption of platform neutrality and was now outdated as platforms mediate online interactions. Critics of the legislation, which included Google and Meta representatives at the forum, defended the current regime as a balanced solution and argue that undermining it threatens freedom of expression, privacy, and intellectual property rights.

Context –  The Marco Civil da Internet was a landmark of open internet legislation. Brazil, under a progressive populist government, undermining the regime, will be an equally important landmark as governments around the world see content online that they don’t like and push platforms to restrict it. China, Russia, Turkey, India, Indonesia and a growing number of states in Central Asia and Africa are easy to criticize as authoritarian online censors. But Germany, Australia and France were early adopters as well. The EU’s Digital Services Act and UK Online Safety Bill claim to set rules for online content moderation while demanding free speech protection at the same time, but they will create similar tensions. The different government authorities simply prioritize different disfavored content.

Top EU Court Rejects Amazon Argument that Italy Cannot Pursue Parallel Antitrust Probe

Report from the Bloomberg

In Brief –  The European Court of Justice has rejected Amazon’s argument that when the European Commission chose to pursue an antitrust complaint targeting its marketplace and logistics policies, that the Italian competition authority was no longer able to pursue its complaint as well. The Italian regulator had initiated an investigation of Amazon in 2019 with an unprecedented focus on logistics practices and fined the company $1.3 billion in 2021. Amazon appealed. The EU filed a similar complaint in 2020 but excluded the Italian market to allow the Italian investigation to proceed. Amazon argued that the EU’s case meant all Member State regulators needed to clear the field. Amazon has since settled with the Commission, agreeing to change how it operates its Buy Box, including displaying a second Buy Box option with a different shipping option, likely slower and less expensive than Amazon’s Prime service. This change is intended to address charges that Amazon uses the Buy Box algorithm to preference sellers who use its FBA logistics business, pushing sellers to adopt fast FBA shipping that carry high fees for Amazon. Amazon’s new commitments should come online in June and will apply in Europe.

Context –  Unlike true marketplaces, Amazon also stores, picks, packs, and delivers the goods supplied by most of their top sellers. The dual role results in combined fees often reaching 50 percent. Amazon also operates unlike other logistics providers, for example storing the products of different sellers of the same products in “commingled bins” across their fulfilment center network, and shipping to a customer from the closest facility regardless of whether the official “seller” supplied the product shipped. It’s unclear how the unique practices of the integrated Amazon logistics and marketplace businesses will eventually play out. Even after settling, Amazon rejected the European Commission complaints but said it hopes its policy Commitments will be a model for its compliance with the new Digital Markets Act (DMA). And overseeing the operation of the new Buy Box is certain to be big test of the DMA enforcement team.

Meta Gets Advice from the Oversight Board on COVID Communications Policy

Report from the Engadget

In Brief –  The Oversight Board has responded to a request from Meta to review its coronavirus misinformation policies by encouraging the company to continue to restrict claims related to the disease and vaccines that public health officials believe pose a serious threat. The company had asked the board to review its July 2022 decision to stop attaching a special COVID Information label to every post mentioning COVID following internal company research showed users who saw the label more than once a month were less likely to click through to the company’s COVID Information Center of authoritative sources. The company also asked the board of free speech policy experts if changes in the nature of the societal response to COVID warranted broader relaxation of the COVID misinformation policies that  saw the company take down 27 million posts on Facebook and Instagram between March 2020 and July 2022. The Oversight Board’s report did not pass judgement on ending the general COVID Information label, but did not back the idea of generally relaxing disease or vaccine misinformation policies. Instead, it called on Meta to continue to regularly reassess, in consultation with public health officials, which claims the company considers false or misleading, and publish a record of when and how it updates that list. Meta’s response was to thank the Board for its recommendations and say that it will continue consulting with experts to help keep people safe.

Context –  The Meta Oversight Board is an innovative attempt to improve the process of platform content moderation without government orders. That effort must be considered in the context of the reality that the job is “impossible to do well” in the words of tech policy expert Mark Masnick. And that there is no correct answer for content in the grey areas of COVID or public health information. Meta has promised to give serious consideration to the Board’s policy recommendations, but not necessarily adopt them. Also, especially after Elon Musk’s purchase of Twitter, major platforms have different policies on things like COVID claims. Is that a problem or a positive?

UK CMA Moves to Block Microsoft’s Acquisition of Activision-Blizzard

Report from the Reuters

In Brief –  The The UK Competition and Markets Authority (CMA) has ruled that Microsoft’s proposed acquisition of videogame giant Activision-Blizzard will undermine competition in the emerging market for cloud-based gaming and therefore is blocking the transaction. The move comes weeks after the antitrust regulator revised its initial objections to the deal by dropping its contention that the Microsoft could undermine competition in console-based gaming by cutting competitors like Sony and Nintendo from popular Activision games. The CMA’s analysis of the console-gaming scenario was methodologically flawed but it continued forward with concerns that Microsoft could use the huge acquisition to dominate the cloud-based videogame market that the CMA contends could become huge in the future by adding to its existing video game, console, software, and cloud services capabilities. Microsoft has said it would appeal the decision.

Context –  The current era of Big Tech antitrust increasingly involves sequential acquisition reviews in the major markets, especially the US, EU, and the UK. Multiple regulators review the same deal and sometimes appear to partner and venue shop in ways that are causing heartburn among business advocates. As the biggest digital acquisition of all time, there was never any doubt that Microsoft’s bid for Activision-Blizzard was going to face reviews in all three markets (and more). Despite claims by chief deal antagonist Sony that Microsoft could use access to top games like Call of Duty to harm PlayStation users, a charge Microsoft always rejected, the emerging cloud gaming market became the chief regulatory concern in the EU and UK. Microsoft has reached a string of 10-year game licensing deals with cloud gaming firms contingent on the deal’s approval and the EU is “market testing” the Microsoft offers. The CMA decision takes a strong stand against those types of “behavioral” remedies and is going the route of just saying no. The move is certain to add to the sense of schizophrenia regarding the UK Government claiming the mantle of digital innovation and activist regulation.

Progressive Democrats Call on Biden Trade Negotiators to Reject Big Tech’s Goals

Report from the Washington Post

In Brief – A group of prominent Democratic Members of Congress led by Senator Elizabeth Warren are urging the Biden Administration to reject business community efforts to use trade talks to weaken or circumvent digital regulations. They are focusing on the Indo-Pacific Economic Framework for Prosperity (IPEF), nascent talks that involve the United States and 13 other countries, including Japan, South Korea, Australia, and India. In a letter to the United States Trade Representative and the Secretary of Commerce, the authors argue that tech companies and their allies will attempt to use trade agreement provisions rejecting discriminatory treatment of digital businesses to undermine a range of legislative and regulatory efforts to address Big Tech abuses. They site examples such as efforts to use USMCA provisions to argue against Canadian legislation to require Google and Meta to pay Canadian media companies for links, the US-South Korea Free Trade Agreement to object to South Korean legislation imposing mandates on the Google and Apple app stores, and arguments that the EU’s Digital Markets Act discriminates against US-based digital giants.

Context –  Digital policy fights keep spilling over into US trade debates. Similar charges emerged last year related to the China competition bill, and in 2019 against the bill implementing the US-Mexico-Canada Agreement. One line of argument is that trade agreements will allow US tech companies to press the US Government to push back on a wide range of foreign digital regulations. A more tortured argument is that trade agreement commitments could tie the hands of the US Congress to enact its own domestic digital regulations in the future. At this point, progressive voices for tech regulation are applauding aggressive legislative and regulatory actions in key foreign markets, especially EU actions on digital gatekeepers, online content moderation, privacy, and potentially AI. As they decry the fact that the US Congress won’t follow suit, they at least do not want the US Government to push back.

Apple Again Largely Prevails Over Epic Games in Federal Court

Report from the Bloomberg

In Brief –  The A three-judge panel of the Federal Ninth Circuit Court of Appeals has affirmed the September 2021 ruling by Federal Judge Yvonne Gonzalez Rogers that Apple is not an illegal monopoly and did not violate federal antitrust law by banning competing app marketplaces on iPhones. The panel criticized some aspects of the initial ruling, in particular the market definition used by Gonzalez Rogers, but ruled that the overall decision was correct based on federal antitrust law. Gonzalez Rogers’ initial ruling was not a complete win for Apple, as she found their “anti-steering” policies prohibiting app developers from informing users about the ability to make purchases off their iPhones to escape Apple’s fees, violated California Unfair Competition Law, and Apple appealed that order. The appeals court upheld the anti-steering order as well. So, again, a federal court result came out largely in Apple’s favor but not entirely. Both companies could consider appeals to the full Ninth Circuit Court of Appeals or the US Supreme Court.

Context –  Epic Games and the global campaign to regulate the Apple and Google mobile ecosystems put together an impressive coalition for Round 2, including contributions from the Biden Department of Justice, 35 State Attorneys General, elite law and business professors, and Microsoft. (Amicus briefs here.) They basically lost again. The most compelling message from the appeals court was that Apple’s privacy and security justification for their walled garden model is strong. And consumers are aware of it. Finally, Epic did not detail an alternative model that met those needs and compensated Apple for their investments. Besides more appeals, which are likely if not hopeful, the top takeaways are that US federal judges are again proving their independence from political leaders, the US and EU digital regulation and antitrust regimes appear to be moving very far apart with the DMA coming online soon, the new in-app payments models of Apple and Google will only become more prevalent and important, and it will be interesting to see how Epic’s parallel trial with Google plays out. While Android has always been more open than Apple’s iOS, it’s less clear how consumers (or a jury) understand it.

Amazon Wins Dismissal of Consumers’ Federal Antitrust Lawsuit on Logistics Practices

Report from Reuters

In Brief –  US District Judge Ricardo Martinez sided with Amazon and dismissed a prospective federal class action complaint brought by two Prime subscribers alleging that the company undermined competition for ecommerce fulfilment services, causing the consumers to pay more for products they purchased on Amazon’s. Martinez ruled that the consumers had failed to show that they had “standing”, meaning the legal right to bring the lawsuit, because they were not themselves the purchasers of the high-priced logistics services sold by Amazon’s Fulfilment by Amazon (FBA) business. Instead, the customers of FBA who may have paid inflated prices were the product sellers. The consumers were “indirect purchasers” of the allegedly monopolized product, and as such are precluded from bringing antitrust suits in federal court.

Context –  Amazon is a unique hybrid of a third-party marketplace and an ecommerce logistics provider. True marketplaces provide a venue for sellers to connect with buyers, but the sellers handle products. But Amazon also handles the products of most top sellers on its platform. Legal challenges targeting this unique combined business have been building because Amazon is both the largest online marketplace and the largest ecommerce fulfilment center business, the combined marketplace and FBA fees charged to sellers often reach 50%, and Amazon aggressively puts products using FBA at the top of shopper search results regardless of cheaper alternatives being available. The European Commission filed an antitrust complaint against Amazon for preferencing sellers who buy logistics services from Amazon, and Amazon agreed to change its Buy Box in Europe to offer buyers a similar product using different, and likely slower and cheaper, logistics. Two federal antitrust suits, and one from the California Attorney General, target similar Amazon practices, and prevailed against Amazon’s motions to dismiss by claiming consumers were directly harmed by Amazon’s “fair pricing” price-parity policy that pushed sellers to raise prices they charged off Amazon to continue making sales on Amazon despite inflated FBA fees.

Google and UK CMA Reach Agreement on Play Store In-App Payments Policy

Report from the Reuters

In Brief –  The Google and the UK’s Competition and Markets Authority have reached a tentative agreement that will end the regulator’s complaint against Google regarding its app store policies requiring app developers use Google’s payments service for in-app payments. Google is proposing to implement in the UK its policies that allow app developers to use approved payments alternatives in addition to the Google service. Developers that choose to use a non-Google service will pay Google 3% lower fees, while those pairing an alternative service with Google’s service will get a 4% fee reduction. These fee reductions are meant to account for the payments processing portion of overall Play Store fees and will be taken off base fees that range from 15 to 30 percent. The CMA is now asking for feedback from market participants.

Context – The most important thing to understand about the global battle over “in-app payments” pitting Google and Apple against app developers is that companies like Epic Games, Spotify, and Match, are not interested in getting to choose their payments processor. They object to high fees from both ecosystem giants. The two have used their payments service to conveniently collect their app fees. We’ve been saying for many months that both will accept payments processing alternatives globally while defending their right to charge app developers commissions and building alternative systems to collect those fees. Apple and Google have been building out new policies and systems in response to challenges in South Korea and the Netherlands and are rolling them out as needed elsewhere. They argues that payments processing is a stand-alone service worth 3 to 4%, and they are not wrong. They also argue that the rest of their commission is in line with a wide range of digital markets. But, again, the big app developers never really cared about the payments processing, they don’t want to pay the fees. They can now give their feedback to the CMA. And we are waiting for a more direct effort at public utility-style price regulation of app store fees.

EU Parliament May Regulate How AI Neural Networks Use Copyrighted Material

Report from Financial Times

In Brief –  As European parliamentarians scramble to pull together their version of a legislative proposal on comprehensive artificial intelligence (AI) regulation, one of the components that it emerging as a major consideration in the wake of ChatGPT’s exploding into a chatbot phenom is the treatment of copyrighted material in generative AI systems. The EU is five years into its effort to enact the AI Act, with versions produced by the European Commission in 2021 and the Council of Member States in 2022. Those proposals are built around a risk-based framework applied to AI uses rather than general models of AI technology. “High Risk” applications face more stringent regulation rather than underlying technologies. Leading EU parliamentarians are now arguing that “generative AI” models themselves should be regulated in addition to the many businesses that could use them in specific applications.

Context –  In a move similar to how more than a thousand very smart people called for a six-month global moratorium in “training” and developing generative AI services, some EU parliamentarians are calling for US President Joseph Biden and EU President Ursula von der Leyen to call a global summit on AI regulation. Just to be clear, neither is happening anytime soon. If ever. AI as a concept is so amorphous and nebulous that even deciding what to regulate is hard. Plus, relatively small firms have made big breakthroughs (think Clearview AI in surveillance) meaning leakage around regulations is a real threat. The issue of copyrighted material in “training data” is interesting. German authors recently jumped on the bandwagon of asking for compensation when AI “learns” by “reading” their online texts. Absent new law, again unlikely, applying the legal concept of fair use to AI training in the US, and the openness of the websites used to gather language examples, is unlikely to mean training AI systems on web content requires compensation. On the other hand, products developed by AI might fall outside Sec. 230 in the US and create a truly interesting liability concern for companies releasing tools to generate material.

Messaging Services Express Continued Opposition to UK Online Safety Bill

Report from the BBC

In Brief – Representatives of eight messaging services that employ strong end-to-end encryption technology, including widely-used Whatsapp and Signal, joined together in an open letter to the UK Government calling for changes to the Online Safety Bill (OSB) to ensure that the government cannot force the messaging services to monitor communications. The companies insist that the OSB gives government regulators the authority to require that messaging platforms used in the country have the capability to scan user messages for illegal content, such as child sexual abuse material (CSAM), which the companies claim is not possible using secure end-to-end encryption that enables only a sender and recipient to unlock a message’s encryption. They argue that the bill is therefore a backdoor effort to prohibit end-to-end encryption. Bill backers argue that the landmark legislation does not “ban” strong encryption, but also argue that messaging platforms must be able to staunch the flow of illegal content. Most of the platforms, including WhatsApp and Signal, have said that they would withdraw from the UK market rather than weaken encryption in manner that they argue would expose all users to increased risks of surveillance and data hacking.

Context – The digital policy fight over encryption stretches back to the 1990’s. It consistently pits civil libertarians and technologists against law enforcement, national security agencies, and victims’ rights advocates looking to extend traditional phone surveillance capabilities to all forms of digital communications. It was a top focus of President Trump’s second Attorney General, William Barr. Globally, the UK and Australia have been similarly aggressive. CSAM and terrorism are often used to justify Internet regulation. That has been the case since the earliest iteration of the OSB. In the US Congress, a parallel debate is underway with the EARN IT Act, which proposes to withdraw Sec. 230 liability protection from platforms that don’t meet certain standards to police objectionable content.

FTC Funding Running into Republican Headwinds (as predicted here)

A PEI Hearing Report

In Brief – A House Energy and Commerce Committee subcommittee hearing on the FY 2024 budget request for the US Federal Trade Commission featured the three current commissioners of the FTC, all Democratic, and contentious questions from the panel Republicans. The independent consumer protection and antitrust enforcement agency has been a central figure in the Biden Administration’s public push for more activist antitrust enforcement, and its budget request calls for a sizeable increase. Led by full committee chair Cathy McMorris Rodgers (R-WA), who said that she would not entertain a budget increase for the agency at this point, Republicans accused the agency and its high-profile Chair Lina Khan of engaging in ideological, legal, and policy overreach. Along with repeatedly criticizing Khan for undermining FTC staff morale, Republicans aggressively questioned high-profile merger reviews, including Illumina-Grail, Meta-Within, and Microsoft-Activision, rulemaking regarding Gig work platforms and franchise business models, and cooperation with the European Commission on the Digital Markets Act that will regulate predominantly US-based digital platforms.

Context – Although Khan, who made her mark as an Amazon critic on antitrust policy while still a law student, had significant Republican support when she was nominated to the FTC in 2021, her tenure as chair has been divisive. But neither she, nor the agency she leads, appears to be pulling back. Several contentious FTC undertakings go well beyond “Big Tech”, including broad new legal thinking on “unfair methods of competition”, an effort to ban employer non-complete clauses, and proposed rulemaking on data privacy (“commercial surveillance”) and Gig work. They are energizing broad-based business community opposition. To be clear, the House Energy & Commerce Committee does not directly fund the FTC, that power rests with the congressional appropriations committees, but those funding bills could end up being a flashpoint later this year as Republicans use them to battle over the progressive FTC agenda.

Arkansas Follows Utah with Social Media Age Limit Bill. It’s 18.

Report from the CNN

In Brief – Following the lead of Utah, Arkansas has become the second state to require parental consent for teens under age 18 to use social media. The state’s Social Media Safety Act requires social media companies to perform “reasonable age verification” when a person in the state sets up an account, which includes checking a user’s government-issued identification or a digitized identification card, and if the user is under 18, they must get parental consent. The bill exempts social media companies with total annual revenues below $100 million, as well as wide range of other online platforms, including those that primarily offer video-gaming, subscription-only services, cloud services, professional networking services, and companies that derive less than 25 percent of their revenue from operating a social media platform. The Act is set to take effect on September 1, 2023, which is six months before the law in Utah, and provides for a private right of action as well as enforcement by the Arkansas Attorney General.

Context – The politics of protecting “kids” from social media, which actually means teenagers, is apparently a winner just about everywhere. US States with undivided Republican control seem especially receptive, or at least quick off the mark. But California, France, and the UK are also prime examples where far more progressive governments are pushing to create a separate, more regulated, and policed version of the internet for teenagers. If there is a “splinternet” coming to Western countries, this is what it looks most like. We don’t have the space here to critique too deeply, but super-smart analyst Mark Masnick does here, here, and here. Creating internet-wide age verification causes most privacy advocates to be very worried. Then there are serious concerns from advocates for at-risk youth who warn of their inability to access online services and support. Finally, actual studies of teen use of social media show it is not generally harmful and is often a positive force. Of course, in the US, there will be constitutional challenges to age-based online restrictions, including this one.

Amazon Knew “Fair Pricing” Policies Pushed Most Sellers to Raise Prices Elsewhere

Report from Bloomberg

In Brief – Newly unsealed filings in California’s antitrust lawsuit against Amazon challenging its “fair pricing” policies revealed that the company knew that most sellers challenged for offering lower prices off Amazon raised their off-Amazon price rather than lowering their on-Amazon price, that Amazon never intended to stop enforcing their price parity policies despite announcing some changes in 2019 after pressure from government officials, and that they were aware that sellers greatly feared the penalty of losing the ability to “win the Buy Box” because of the detrimental effect that had on sales on Amazon. In March, California Superior Court Judge Ethan Schulman rejected Amazon’s efforts to have the complaint brought by California Attorney General Rob Bonta dismissed. The California AG argues that Amazon penalizes sellers who offer lower prices on alternative internet sites to blunt competition from alternative marketplaces that charge lower fees. Consumers are harmed because sellers raise their prices on other sites to match those that are elevated on Amazon to account for higher marketplace and logistics fees. The ecommerce giant argues that their policies benefit consumers by pressing for the lowest prices to be offered on the Amazon Marketplace.

Context – Amazon’s price parity policies first drew the attention of antitrust officials in Europe, with the company agreeing to make changes in 2013. Besides the recent revelations that company officials never intended to abandon their price enforcement practices, it is interesting that Amazon’s recent antitrust settlement in Europe, which included changes to how it awards the “Buy Box,” did not include a price parity offer. In the US, antitrust suits alleging illegal price parity conduct are adding up. The California suit joins two similar complaints in federal courts in the State of Washington that have survived motions to dismiss. Amazon’s especially large share of the online marketplace market in the US, and the fact that average fees on top sellers are reaching 50 percent, are clearly adding to their legal vulnerability.

Montana State Legislature Passes Law Banning TikTok

Report from the Washington Post

In Brief –  The Montana state legislature has passed legislation to ban TikTok in the state, becoming the first US state to do so. If Governor Greg Gianforte (R) signs the measure, it would become illegal to download TikTok. The bill imposes penalties of up to $10,000 a day for any entity, such as the Apple and Google app stores or TikTok itself, that makes the popular short-video social media app available to residents. The ban would not go into effect until January 1, 2024, and lawsuits challenging its constitutionality, especially on First Amendment grounds, are certain. Like the Trump Administration’s efforts in 2020, and the Biden Administration more recently, the Montana bill says that if the service is sold to a company that is not based in an “adversarial nation,” the ban would not take effect.

Context – Once again, it’s important to remember what happened when President Trump banned TikTok (and Chinese “super app” WeChat). One irony of the Chinese app fight is that banning them is justified in part by the fact that there is no rule of law or independent judiciary in China to limit the power of the government to manipulate the TikTok algorithms or access user data. But Chinese-based companies can access truly independent US courts. And the last time around federal judges were very sympathetic to lawsuits challenging the TikTok (and WeChat) bans. Arguments were based on the First Amendment and a Cold War era federal law called the Berman Amendment. Then Trump lost the election, and President Biden called off the bans and court battles. Now, a revived ban, whether at the federal or state level, is certain to restart the very interesting (and likely lengthy) legal battles. Over the ensuing years, TikTok’s popularity has grown immensely, which is resonating with Democratic Party officials. In addition, the Chinese Government cracked down on its digital giants (consider Jack Ma’s travails) destroying claims that such companies can operate independently. And repeated leaks from inside TikTok have exposed ByteDance interference that raised questions about whether any of the company’s plans can be trusted.

Talk of US Federal Government AI Regulation… It’s Not Happening

Report from Reuters

In Brief – As the public profile of Al-enabled ChatGPT has exploded in recent months, a handful of federal agencies are stepping into the void of direct regulation of AI technology in the United States with messages indicating that they are engaged. The National Telecommunications and Information Administration, an agency of the US Department of Commerce, has announced that they are soliciting public comments on how an AI auditing regime could assess whether AI systems include harmful bias or distort communications to spread misinformation or disinformation. This follows messages from the Federal Trade Commission (FTC) and the US Department of Justice’s Antitrust Division that they will be focusing attention on how big companies might use AI-tools in anticompetitive pursuits. Finally, the FTC’s Consumer Protection Bureau has said they are on the lookout for deceptive claims about AI services and tools.

Context – When you read about the government rushing in to regulate Artificial Intelligence, take a deep breath and keep a few things in mind. First, AI is such an amorphous, nebulous concept, that even deciding what it is, is very difficult. Second, consider the jurisdiction. In this case, the topic is US Government regulation. No matter the hyperventilating, Congress is not about to regulate general AI technology. And Biden Administration regulators will need to get their stories straight. In a moment of ChatGPT-inspired irony and humor, as many government officials worry about AI chatbot-search hybrids going off the rails and defaming people, US Department of Justice litigators argued that Google’s alleged search monopoly kept ChatGPT and other AI search tools off the market for years, as if things would have been better with earlier models proliferating online. Yes, Europe will eventually wrap up their massive AI Act legislation. But ChatGPT’s super chatbot is slowing things down there by causing some European Parliament leaders to rethink their whole risk-based framework regulating AI applications not tech.

Europe’s Reaction to ChatGPT — Start Regulating People

Report from the Wall Street Journal

In Brief –  Five years into the EU’s effort to establish a comprehensive legal scheme to govern Artificial Intelligence, leading members of the Parliament are proposing to change the regime in order to regulate AI-enabled tools like ChatGPT. The AI Act versions from the European Commission (2021) and the EU Council of Member States (2022) are built around a risk-based framework applied to AI uses rather than general models of AI technology. The most stringent regulations apply to “high risk” uses that threaten fundamental rights. General AI techniques, including “generative AI” services like ChatGPT, would not be regulated like the applications that use them. Despite many months of work to arrive at a parliamentary position on the AI Act so that the three EU governing bodies can negotiate a final version, leading parliamentarians are now saying that the regime should be changed to more stringently oversee the general AI tech models and impose more duties and liability on tech developers like OpenAI rather than just the firms that use the AI tools in specific applications that are released to the public.

Context – The EU has been working to develop what proponents regularly describe as the global model for smart AI regulation. So their reaction to ChatGPT brings to mind the saying that “No plan survives first contact with the enemy” and the culminating scene from the Thomas Crown Affair (1999) when the NYPD response is to “Start arresting people.” The Italian data protection authority was first on the scene, banning ChatGPT in Italy for suspected violations of the GDPR. Spain’s authority announced similar concerns and the Bloc’s privacy regulators are now engaged. There is not enough space here to explain how “large language models” don’t maintain databases on people like traditional data services, but try asking ChatGPT about yourself and see what you get. In reaction to Italy’s ban, EU tech policy leader Margrethe Vestager has said she still supported regulating AI uses, not policing underlying technology. And any meaningful shift in the Parliament is likely to delay what has already been a long, complicated, and contentious effort.

Pro-Biden “Dark Money” Group Running Ads Touting President’s Attack on Social Media

Report from TechCrunch

In Brief – The UK Competition and Markets Authority has announced that it is beginning a formal investigation of the competition effects of Amazon’s proposed $1.7 billion acquisition of iRobot, the manufacturer of the Roomba robotic vacuum devices. The UK regulator now joins the European Commission Competition Authority and the US Federal Trade Commission in probing the acquisition, although no competition agency has formally moved to block the deal. In February, reports surfaced that the European Commission was continuing to proceed toward a formal investigation that would focus on privacy and data collection, in particular concerns that the robotic vacuums collect sensitive in-home images, while the company claims they have only basic room-mapping technology, and a company spokesperson reiterated that “We’re working cooperatively with the relevant regulators in their review of the merger”.

Context – Amazon’s bid for iRobot had a regulatory bullseye on it from the start. However, Roomba’s declining market share and the company’s lack of profitability raise questions about the strength of a legal challenge. Big Tech critics quickly raised a wide range of concerns, including traditional antitrust charges that Amazon could leverage their online commerce dominance to undermine the large host of robot-vacuum competitors and reinforce their own smart-home product suite. A UK-based consumer group urged the CMA to step in based on these grounds. Then there were the less traditional privacy-based charges regarding in-house images, but they are said to be moving the EU regulators. And similar accusations recently surfaced about in-car videos and Tesla. The US FTC, led by Lina Khan who rose to prominence as a progressive antitrust reformer challenging Amazon’s business model, declined to challenge Amazon’s larger MGM and One Medical acquisitions, but it continues to probe the iRobot deal and reports indicate that the manufacturer’s residual (though declining) market share of 62% might make FTC litigators more comfortable.

Led By Italy, More EU Privacy Regulators Looking at Banning ChatGPT Operations

Report from the Reuters

In Brief –  The decision of the Italian data protection authority, also known as Garante, to temporarily ban the operation of OpenAI’s chatbot phenom, ChatGPT, has inspired data protection regulators in Germany, Ireland, and France to publicly consider stepping forward to regulate the AI-enabled digital service as well. There is no specific regulatory scheme for Artificial Intelligence technology in Europe, and the Italian regulators have stepped into the fray based on existing laws, in particular the General Data Protection Regulation (GDPR) that governs how companies handle personal information. Garante has raised a number of concerns regarding ChatGPT, including the legal basis of OpenAI’s data collection to build its “large language model”, the prospect that the service produces inaccurate information about individuals, the inability of the company to prohibit children under age 13 from using a tool the regulator says is potentially dangerous, and the risk of data breaches revealing user personal information, such as occurred in March.

Context – While European data protection regulators and privacy advocates call for government action to intervene with ChatGPT and other similar AI-enabled services, other government officials have been more reticent to call for action. For example, Italy’s Deputy Prime Minister has criticized Garante for banning the chatbot, and a German government spokesperson said a ban in that country would not be needed. EU Commissioner Margrethe Vestager, a leader of the Commission on digital policy, has called for applying EU laws and regulations equally regardless of what technology systems operate behind the scenes. Despite not moving fast enough for some, the EU is a couple of years into developing its AI Act legislative package that includes an AI Liability Directive and AI updates to the EU Product Liability Directive. Oh the other hand, the Biden Administration has continued the US policy of sticking to high-level principles and policy goals, while the UK Government has highlighted what it calls a lighter touch AI oversight plan.

Pro-Biden “Dark Money” Group Running Ads Touting President’s Attack on Social Media

Report from Washington Post

In Brief –  Future Forward USA Action, a pro-Biden “dark money” political group with links to prominent Silicon Valley Democrats, has mounted a major ad campaign touting his State of the Union (SOTU) call for Congress to unite against the tech giants. The group has reportedly spent over $900,000 on TV ads in recent weeks promoting his remarks during February’s State of the Union Address urging lawmakers to “finally hold social media companies accountable.” The 15-second ad displays the logos or icons of tech companies including Snapchat, Twitter, Facebook and TikTok, and urge viewers to call the White House to thank Biden “for delivering for families.” Advocates for increased regulation of the digital platforms claim that the advertising indicates that Big Tech will be an increasingly visible campaign issue.

Context –  The Biden Administration likes publicizing its anti-Big Tech agenda. While the President only spent a few lines in his SOTU criticizing tech giants, it was enough content for political ads. In campaign mode last fall, the White House boosted progressive allies promoting action on antitrust, privacy, Sec. 230, and social media algorithms. It all fell short. On one hand, the prospects for legislation aren’t better in 2023 or 2024 than they were the past two years. Most of what’s changed in the political landscape reduces prospects for bipartisan tech legislation in Congress. But it is worth considering the longer-term effect of pummeling social media as a top Presidential campaign issue for 18 months. There is no way Republicans will let themselves be rhetorically outflanked. Beating up social media giants is already a top conservative issue. In Utah, Republicans just passed a law that includes requiring parental approval of social media use by teens under age 18, special parental access passwords, and default blocking from 10:30 pm to 6:30 am. While the federal courts, especially the Supreme Court, remain the most likely nexus for major federal policy change in the US, a top tier national political fight to be tougher on social media could have unforeseen impacts further down the line.

Google Wins Some Relief from Indian Legal Panel on Android Antitrust Orders

Report from the Reuters

In Brief –  Although India’s National Company Law Appellate Tribunal (NCLAT) has upheld the $161 million penalty imposed on Google by the Competition Commission of India (CCI) for anticompetitive Android practices, the panel overturned four of the ten CCI conduct remedies. In January, the Indian Supreme Court refused Google’s appeal to suspend any of the CCI antitrust order, but the high court ordered the NCLAT to hear the case and issue a ruling by the end of March. Google will now not need to allow third-party app stores to be offered from inside its Play Store, permit users to remove pre-installed apps such as Google Maps, Gmail, and YouTube, or end all curbs on so-called “side-loading”. Following the Supreme Court’s January decision, Google announced a number of significant changes to Android policies in India, including allowing device makers to license individual apps for pre-installation, giving users the option to choose their default search engine, and establishing a process for app developers to employ alternative’s to Google’s payment service for in-app payments.

Context –  Although Apple and Google are both under fire for their mobile ecosystem practices, they follow very different business models. Apple’s “walled garden” means they make all the devices, do not license the operating system, and strictly control all apps. Google’s Android is a kind of hybrid. Android source code can be used in an “open source” manner, but manufacturers who produce devices they brand as “Android” (think green robot logo) must follow many contractual mandates, including regarding apps, for all their devices. While some of Google’s mandates were successfully challenged by the EU in 2018, regulators in South Korea and India, two markets where Android devices hold particularly large market shares, have pushed for further changes to the Google business model. On the key question of fees collected on in-app payments, Google is again proposing a 4% discount to developers using a payments alternative in India, a fee cut that appears not be close to what app developers such as Epic Games want.

Federal Judge Rules Former GrubHub Driver was an Employee Back in 2015-2016

Report from Bloomberg

In Brief –  In the latest ruling in a Gig work litigation battle stretching back to 2016, Federal District Court Judge Jacqueline Scott Corley, who has overseen the case throughout its tortuous path, determined that driver Raef Lawson was serving as a GrubHub employee in late 2015 and early 2016, not an independent contractor, at least for wage and overtime claims. The judge, who ruled in 2018 that Lawson was an independent contractor but noted that California legislators might consider Gig work issues, subsequently saw California’s top court establish the “ABC Test” for worker classification in the 2018 Dynamex decision, the state legislature enact AB 5 fully instituting the Dynamex decision in 2019, and then have state voters exempt Gig delivery and ridesharing platforms from the AB 5 regime in 2020. A series of other court decisions determined that the strict ABC Test standards should be applied retroactively, which overturned Lawson’s initial loss, but that Prop. 22 exempting GrubHub and similar platforms from the ABC Test regime was not retroactive. Last September, Judge Corley ruled that the ABC Test doesn’t apply to Lawson’s claim for business expense reimbursement. However, wage and hour claims are covered. The judge determined that Lawson is entitled to $65.11.

Context –  The sense that California would lead the way in getting rid of Gig work seems forever ago. This case is like a living history lesson. Gig workers as contractors. Dynamex. ABC Test. AB 5. “Blue” California voters solidly backing Prop. 22. Federal bills on Gig workers stalled even when Democrats controlled the White House and Congress. Gig worker classification bills not enacted in Blue States either. The Biden Administration Labor Department has proposed a new rule changing the criteria that businesses are advised to use when classifying a worker as an employee or a contractor, but it has not been implemented or enforced. Court challenges are expected, and similar worker classification efforts by regulators without legislative backing have suffered setbacks in the past. And yes, the award was for $65.

Meta Considering Banning Political Ads in Europe Due to Transparency Mandates

Report from the Financial Times

In Brief –  Meta is reported to be considering banning all political ads on its services in Europe in response to regulations that would require the company to track information and report information on who is buying ads, who they are attempting to target, how many users saw an ad, and how much the advertiser paid to the platform. The European Commission, Parliament, and the Council of member states are aiming to wrap up the new regulation by June 5. There appears to be agreement over the Commission’s description of political advertising as the promotion of a message by a “political actor or which is liable to influence the outcome of an election”. However, disagreement remains over how much targeting should be permitted, with some defending ad targeting as valuable to smaller political parties, while others oppose highly targeted ads in all contexts. Meta has said that political ads are a small portion of its business and the content often leads to unpleasant user experiences anyhow.

Context –  The debate over targeted political advertising is a microcosm of the broader, and seemingly endless, debate over targeted digital advertising in general. Many privacy advocates oppose targeted advertising. They seem to believe that effective advertising connecting people who want things, with people trying to sell the same things, is bad. But consumers have generally preferred ad-based online business models rather than paying directly for online services, and smaller businesses, like smaller political parties, benefit most from efficient, highly targeted, ads. Then you can add repeated controversies over supposed misinformation on political issues and ads, and the extreme sensitivity of elected officials regarding ads for and against themselves. As more national governments threaten to force Meta to pay media companies when users, including media companies, post articles, Meta is threatening to simply ban all news articles, saying users want TikTok-like entertainment, not controversial news, anyhow. Maybe they will do likewise with political ads.

Spain’s Competition Authority Investigating Google on Media Payments Policies

Report from TechCrunch

In Brief –  Spain’s competition authority, the CNMC, has announced that it is opening an investigation of Google’s practices related to the licensing of news content from Spanish media publishers. The regulator is concerned that the digital giant is engaged in an abuse of a dominant position vis-à-vis the publishers and allegations that it is imposing unfair commercial conditions on them to exploit content protected by intellectual property rights. The CNMC is acting in response to a complaint by the Spanish Center for Reprographic Rights (aka, Centro Español de Derechos Reprográficos or CEDRO).

Context –  Newspapers have been complaining for two decades that the internet ruined their business model and that Google and Facebook, the two largest digital advertising platforms, should pay them more money. Government in France and Australia have most aggressively come to the aid of media companies, but the effort to force Google and Meta to pay up is a global phenomenon. Australia’s 2021 showdown with the two digital giants resulted in a law that pushed the companies to negotiate payments with many Australian media companies, backed by a threat of government-led arbitration. The digital giants both drew lines in the sand, with Meta rejecting being forced to pay for news media content posted by users, and Google rejecting payments for simple search links. Canada is the latest country to propose following a version of the Australian model, and Meta is again threatening to block user posted news and Google has tested a no-Canadian news version of search in Canada. The issue has evolved somewhat differently in Europe, with the EU amending its copyright law in 2019 creating “neighboring rights” aimed at Google news “snippets”. France’s competition authority took the lead pressing Google to pay media companies under a mandatory licensing regime. Google and the French regulator resolved a two-year legal battle last summer, Google announced an expansion of its curated news services that pays many media companies, and also the search giant also reopened Google News in Spain after an eight-year payments dispute.