Archive – 15 Sep 2023

French Finance Minister Says Europe Needs to Prepare to Fully Deploy Digital Taxes

Report from Bloomberg

In Brief – French Finance Minister Bruno Le Maire has publicly declared that global tax reform is being “blocked” by countries including the United States, Saudi Arabia, and India, and that Europe should prepare to move forward on new taxes targeting large digital platforms. The preliminary tax reform deal negotiated at the OECD, which achieved near consensus support in 2021, attempts to address two long-running tax concerns – the right way to tax digital giants like Google and Apple, and frustration that some countries offer “tax havens” to multinational companies to win investment. The first part of the deal, dubbed “Pillar 1”, was initially a new tax on approximately 20 very large, mostly US-based, digital companies, but was modified to cover 100 or so highly profitable consumer-facing businesses, allowing governments in countries with large numbers of consumers to expand their ability to impose taxes on the consumer-facing companies based in other jurisdictions. The second half, “Pillar 2”, aims to undermine tax havens by having countries agree to tax multinational companies at least 15 percent. Despite real progress in 2021, the campaign has hit hurdles in the US, Europe, and with some developing countries.

Context – The Biden Administration unlocked long-stalled global tax reform talks by shifting the US focus from simply opposing new taxes on digital companies to supporting the 15% “global minimum tax”. But that progress seems like many years ago. Instead, it feels more like 2020, with US digital companies facing discriminatory national digital services taxes (DSTs) in a growing number of countries. France led the pack in 2020, but other European countries, India, and Turkey followed on. The Trump Administration strategy was to block DSTs by threatening trade retaliation. Along with achieving the OECD progress, the Biden Administration reached agreements with several European countries and India, who each started collecting national DSTs in 2022, to agree to rebate tax overages back to the digital companies once the full global tax deal went into effect. But those deals expire in 2023.

US Department of Justice Prepping Another Google Antitrust Suit — Map Services

Report from the Bloomberg

In Brief – The Department of Justice (DoJ) is reportedly ramping up its antitrust probe into Google’s Maps service, re-interviewing potential witnesses among competitors and customers, with an aim of filing a federal antitrust lawsuit this year. Maps is one of Google’s most widely used services, with the company reporting that more than a billion people use it every month and more than 5 million active apps and websites are using Google Maps Platform core products every week. The DoJ is reported to be focusing on the terms and conditions that Google imposes on app and web developers that use technical capabilities from Google Maps, such as location search, in particular prohibiting developers from combining some Google mapping services, such as location data, with map services from rivals, as well as how Google bundles Maps into its Google Automotive Services infotainment system service offered to automakers. The company, who has faced criticism over its Maps practices from regulators and others in the past, has said that its policies are designed to improve user experiences, minimize errors that can lead to safety risks, comply with license terms from location data suppliers it must adhere to, and that developers are free to use mapping systems from other suppliers.

Context – Google’s Maps policies are emerging as the most popular target of antitrust scrutiny after the three pillars of Google antitrust investigations outlined by the European Commission nearly a decade ago – Search, Android, and Advertising. The DoJ, now led on antitrust matters by a long-time Google nemesis, already has two antitrust suits targeting Google underway, one alleging anticompetitive search practices and one targeting Google’s alleged digital advertising services monopoly. Europe has been active on Maps as well. Germany’s national competition authority opened an investigation of Google’s Maps Platform last June that directly parallels this DoJ investigation, and the Italian competition authority fined Google in 2021 for anticompetitive practices on their auto services platform.

Amazon’s One Medical Deal Closes After FTC Says It Won’t Sue to Block it (Yet)

Report from Reuters

In Brief – Amazon has completed its $3.9 billion acquisition of One Medical following the Federal Trade Commission (FTC) notifying the companies that it would not be sue to block the deal prior to its completion. However, the agency did make clear that it has not formally cleared the acquisition, will continue its review, and retains the right to challenge it in the future. One Medical operates a network of in-person medical offices and also provides virtual medical services to patients paying a $199 annual membership. Amazon will gain health facilities, professional medical staff, technology that enables virtual doctor visits, and medical data on more than 700,000 patients. Progressive critics quickly raised concerns with the deal arguing that the company already has “too much economic power” and that there were unique risks related to Amazon gaining personal health data. An agency spokesperson said that the FTC “will continue to look at possible harms to competition created by this merger as well as possible harms to consumers that may result from Amazon’s control and use of sensitive consumer health information held by One Medical” and ongoing agency investigation of the acquisition could be incorporated into its broad competition investigation of Amazon’s many businesses.

Context – Antitrust regulators in many major markets have long expressed concerns with the acquisition activities of the largest digital platforms, but rejections of specific deals have been very limited. FTC Chair Lina Khan, who rose to prominence as a progressive champion of aggressive antitrust enforcement by focusing on Amazon’s unique ecommerce business model, is a notable example. Amazon has certainly offered Khan opportunities to challenge deals, including MGM, One Medical, and iRobot, but each represented tough challenges under traditional antitrust analysis. The FTC has now passed on challenging the first two, although it is worth noting that there are reports that the European Commission is increasingly concerned about potential data collection implications of Amazon’s iRobot acquisition.

Antitrust Stalwart David Cicillini Unexpectedly Announces Departure from US House

Report from the New York Times

In Brief – US Rep. David Cicilline (D-RI), a tenacious champion of reforming US antitrust laws to reign in the largest digital platforms, has unexpectedly announced that he will resign his House seat on June 1 and become the CEO of the Rhode Island Foundation, a major funder of community service non-profits across his home state. An influential progressive leader in the House who was entering his seventh term, Cicilline emerged as a leader on digital policy issues during his two terms as the Chairman of the House Judiciary Committee’s Antitrust Subcommittee. He developed a close working relationship with his Republican counterpart, conservative Rep. Ken Buck (R-CO), and put together a landmark set of bipartisan antitrust bills focused on a range of Big Tech practices that were the subject of a major lobbying effort through the end of the 117^th Congress.

Context – Cicilline played a huge leadership role in turning the progressive antitrust reform movement into a substantive legislative campaign. One reason the effort was considered a credible threat to the tech giants was that it brought together Progressive Democrats who long called-for major antitrust reform and Republicans angered by digital giants that they considered ideological opponents. Cicilline’s work with Buck was the clearest manifestation of that bipartisan coalition. The House Judiciary Committee reported out the bills backed by Cicilline and Buck in June of 2021, and the Senate Judiciary Committee passed similar bills in January 2022. Despite repeated claims by backers that they “had the votes”, growing Republican concerns with regulatory overreach, especially at the FTC, combined with Democratic naysayers from California and moderate districts, held the bills back. Last November’s results meant that Cicilline lost his chairmanship, and then Republican leaders bypassed Buck for the subcommittee helm, installing an antitrust skeptic and reinforcing the impression that major antitrust reform was off the agenda for 2023. We expect to see both running for higher offer before too long.

Social Media Regulation for Teen Users Gets Out of the Gate with US Senate Hearing

Report from CNN

In Brief – Advocates of regulating how social media sites serve younger users, especially teenagers, restarted their 2023 congressional efforts with a hearing in the Senate Judiciary Committee, calling popular services such as Instagram, TikTok, and YouTube addicting and dangerous. During the recently ended 117th Congress, legislation to impose new duties and restrictions on digital platforms serving younger users generated significant bipartisan support, especially the Kids Online Safety Act and the Children and Teens’ Online Privacy Protection Act in the US Senate. While both bills were reported out of the Senate Commerce Committee for consideration by the full body, neither was considered on the Senate floor as disagreements with the House of Representatives over privacy policy priorities stymied action through year’s end.

Context – The seemingly strong bipartisan support for overhauling how social media platforms serve millions of US teens generates plenty of attention but doesn’t cross the finish line. But the backers keep trying. The politics are unquestionably good, and the backers are likely sincere believers. While most astute analysts believe that major legislation, including on most digital policy issues, will face bigger hurdles in 2023 due to the newly divided Congress, a long-time Senate champion of tech reforms recently called out two issues, new rules to protect kids online, and dealing with technology threats from China, as the best chances for bipartisan success. “Protecting kids” has always been the top justification of tech regulation, and the UK, France, and California are all pushing towards a kind of age-based “splinternet” that could soon require some manner of internet-wide age verification. President Biden again mentioned protecting kids online in his State of the Union Address. And states such as Maryland are considering their own California-style regulations despite many progressives being deeply concerned that regulating how younger people access and use the internet could threaten many marginalized and at-risk teens.

Amazon Acknowledges Tesla Accused of Firing Workers to Disrupt Union Drive at Buffalo Facility

Report from the Bloomberg

In Brief – Labor organizers in a Tesla facility in Buffalo, New York, are claiming that the electric car giant led by publicly anti-union Elon Musk has fired dozens of employees to stymie their union drive and intimidate workers. The organizing effort is being led by the same Service Employees International Union affiliate that kicked off a successful Starbucks labor campaign in a Buffalo café just miles from the Tesla plant. The Tesla workforce in Buffalo includes more than 800 “Autopilot analysts”, a non-technical role with a starting pay of around $19 per hour that involves helping “train” image-recognition “neural networks” by reviewing imagines that Tesla vehicle sensors capture and label to determine the accuracy of the automated systems and correct errors to improve their accuracy. The labor organizers are asking for increased pay, better job security, reduced digital key-stroke monitoring, and eased production quotas for the Autopilot analysts, and have filed a complaint with the National Labor Relations Board claiming that the employees were illegally terminated to discourage union activity.

Context – The highly politicized topic of unions organizing inside Big Tech increasingly requires filtering for “tech” or “not tech” workers. The most visible recent activity has been at video game companies, with labor wins at two Activision Blizzard studios and Microsoft-owned Zenimax. But they were limited to “game testers”, who tend to be hourly employees earning around $20 per hour, not programmers or developers. That worker profile is a lot like the Tesla image analysts. Chalk them up with unions representing Amazon fulfilment center workers and Apple retail employees. Big Tech companies but “not tech” labor. At Proletariat, a small videogame studio in Boston, a recent effort to unionize the full staff, including developers and programmers, fizzled after early claims of success. Instead, the biggest win among skilled digital workers was last year at the New York Times, when 600 employees, including engineers and designers, voted to organize in defiance of the normally pro-labor Times management.

The Supreme Court Considers Terrorism Liability for Communications Platforms — Twitter v Taamneh

A Report from Platform Economy Insights

In Brief – For the second day running, the Supreme Court heard extensive oral arguments in a case raising the prospect that digital platforms can be held liable for harm caused by terror groups. Yesterday was Gonzalez v Google, which stemmed from a 2015 ISIS attack in Paris. It focused on the scope of Section 230. Today was Twitter v Taamneh, which arose from an ISIS attack in Istanbul in 2017, and involves whether an internet communications platform can be found to be aiding and abetting a terrorist enterprise under the Anti-Terrorism Act, and therefore be liable for the group’s attacks, for providing its general services to members of the terror organization. Google had prevailed in the lower court based on the argument that whatever videos got through its efforts on YouTube to block pro-terrorism communications, Sec. 230 protected it from liability. However, Gonzalez made the case that Google’s recommendation algorithms and other automated systems were implicated in promoting ISIS content, that those actions were outside the scope of Sec. 230, and the Biden Administration backed that limited view of Sec. 230. The lower courts in the Twitter case, which also involving social media and ISIS, followed a different line of reasoning, arguing that the Anti-Terrorism Act conferred liability on Twitter not based on the content of ISIS communications, but instead on the fact that ISIS members used Twitter’s services in general. Therefore, the court ruled that Sec. 230 did not apply. Twitter’s appeal argued that it was not liable for harms from the Istanbul attack because there is no evidence its services were used in the attack, nor that it had any knowledge that its services were used by the people who planned or carried out the attack. Unlike yesterday, where the Biden DoJ was arguing against Google’s interpretation of Sec. 230, the DoJ sides with Twitter and argued that general commercial services, in particular those that operate remotely, should not be liable for aiding or abetting terrorism or other criminal activity when their service is just generally used by people who engage in criminal activity absent specific knowledge of the illegal activity.

Key Links – The oral argument itself is available here. The attorney representing Twitter is questioned for an hour. The Biden/DoJ attorney starts at 1 hour, 1 min. Taamneh’s lawyer begins at 1 hour, 51 mins. Finally, there is a short rebuttal at 2 hours, 25 minutes provided to the Twitter lawyer. A summary of the case and downloadable copies of all the briefs are available here.

A Few High-Level Thoughts –

• Once again, they are interesting to listen to, but it’s not smart to think you can ascertain the “winner” from Supreme Court oral arguments. In the context of these two cases it’s especially tricky because there is enough overlap that the justices have more options than usual to craft resolutions. We will have to wait for the decisions this summer.
• Yesterday’s case and oral arguments were a lot more focused on “the internet” due to the specific question related to Sec. 230, one of the absolute landmarks of internet law and policy. Today was more about how to apply a potentially very expansive concept of secondary liability for terrorism to a remote commercial service. All parties understood that the same rules would be applying to a range of consumer services.
• Justice Kagan was focused on the prospect of one standard for banks, who are sometimes held liable when they provide services to criminal enterprises, and a lesser one for social media platforms. She argued that there is no need to show that a bank’s services funded or contributed to a specific criminal act, so why shouldn’t offering communications tools to a terrorist group be the same?
• Yesterday it was hard to identify major ideological or partisan camps among the justices. Today, a number of the conservative justices appeared concerned about the prospects of meaningfully expanding aiding and abetting liability risks to businesses who are not alleged to have any direct connection to a terrorist or criminal act.
• One of the toughest lines of questioning for the internet platform was again what the law should do about platforms that consciously don’t do the right thing. Yesterday, it was whether Sec. 230 protects an internet service that sets its algorithms to handle third-party content in a harmful manner. Today it was whether a social media company that consciously disregarded warnings that terrorists used its service should be liable for that group’s attacks.
• Reflecting on the five-and-half hours of oral arguments total, a top takeaway was that while all the justices appear uncomfortable with the idea that digital platforms might be able to exploit Sec. 230 to get a free pass on bad behavior, the fact sets of these two cases don’t raise that scenario.
• In that light, one route might be for the justices to try to find an out like the scenario mentioned by Justice Barrett yesterday, when she raised the idea that if Twitter (and the US DoJ) prevailed, then maybe the High Court could dispense with the Gonzalez v Google case on those grounds and avoid ruling on Sec. 230 and algorithmic ranking. And presumably wait for a Sec. 230 case with a platform that behaves badly.
• A second route, and one that likely worries many advocates of digital platforms more, would be for the justices to try to craft some manner of new Sec. 230 conditions or tests that would try to protect the use of algorithms or automated services that they adjudge general or benign, but not algorithms that are designed to do “bad” things.
• And face it, we are all waiting with bated breath for the court to accept an appeal involving either the Texas or Florida social media laws. Or both. Justices Thomas, Alito, and Gorsuch have said they want to. These terrorism liability cases are but hors d’oeuvres to that main course.

The Supreme Court Considers the Scope of Sec. 230 — Gonzalez v Google Oral Arguments

Report from Platform Economy Insights

In Brief – The US Supreme Court heard nearly three hours of debate in the oral argument for the case of Gonzalez v Google, which is the first time that the High Court has taken a case testing the scope of Section 230 of the Communications Decency Act, the landmark law that largely protects internet platforms from being civilly liable for content created by third parties. The case in question involves the family of an American victim of the 2015 Paris ISIS attack who sued Google’s YouTube service for sometimes recommending pro-ISIS videos and therefore should be partially liable for the harms of the attack. The key issue in the case is whether actions of internet platforms to organize, rank, recommend, order, or otherwise present third-party content is outside the scope of Sec. 230. Of note, the Biden Administration’s Department of Justice participated in the debate making that argument, which is mostly in line with Gonzalez and against Google and most of the internet business community.

Key Links – The oral argument itself is available, somewhat ironically, here on YouTube. Audio actually starts around minute 20. The attorney representing Gonzalez is questioned for over an hour. The Biden/DoJ attorney starts at 1 hour, 27 mins. Google’s lawyer begins at 2 hours, 14 mins. Finally, there is a short rebuttal at 2:57 provided to the Gonzalez attorney. For those who prefer to read, a summary of the case and downloadable copies of all the briefs, including the many amicus briefs, are available here. Supreme Court amicus briefs, with their relatively tight space limits, are often excellent legal and policy issue reading.

A Few High-Level Thoughts – The oral argument itself is available, somewhat ironically, here on YouTube. Audio actually starts around minute 20. The attorney representing Gonzalez is questioned for over an hour. The Biden/DoJ attorney starts at 1 hour, 27 mins. Google’s lawyer begins at 2 hours, 14 mins. Finally, there is a short rebuttal at 2:57 provided to the Gonzalez attorney. For those who prefer to read, a summary of the case and downloadable copies of all the briefs, including the many amicus briefs, are available here. Supreme Court amicus briefs, with their relatively tight space limits, are often excellent legal and policy issue reading.

• It’s a rookie mistake to think there is a straight line from oral arguments to a decision. Justices have many interests and motivations behind what they ask. It’s fun to speculate and predict, but We have to wait for the decision this summer.
• This week’s consideration of two big “Internet” cases is unique. Today, Gonzalez-Google was focused on the scope of Sec. 230. Tomorrow is a similar Internet service case, Twitter v Taamneh. It also regards potential platform liability for an ISIS terror attack. But it is technically focused on the scope the Anti-Terrorism Act to general online services. A “normal” oral argument is one hour long… we already have three hours in one day!
• Yes, the Justices always scrutinize the lawyers on both sides, but the lawyer representing the Gonzalez family faced very tough questions for well over an hour. A number of Justices commented that they did not understand the argument that he was making. It did not seem to go well for him.
• There was a lot of conflation between the Google/Sec. 230 case and tomorrow’s Twitter case. That lack of clarity from the justices threw off the discussion at times.
• As any PEI reader knows, the partisan divisions on Sec. 230 and platform liability are clear, with Democratic officials generally arguing that platforms are not aggressive enough in limiting access to harmful content. Justice Jackson was especially focused on the idea that Sec. 230 itself was really intended only to facilitate the ability of platforms to take down objectionable content, not otherwise promote content.
• Justice Kavanaugh and Chief Justice Roberts seemed especially reticent to have the Court dramatically change the way Sec. 230 is applied given the social and economic impacts of the internet. They were saying that it was probably best to leave such change to Congress.
• Justice Barrett asked about whether individuals who re-tweeted something legally problematic might be in jeopardy by falling outside Sec. 230 by recommending something, a line of questions that the internet companies must have liked.
• During the examination of Google’s attorney, the prospect that biased algorithms, or algorithms that intentionally recommended bad things, would also receive Sec. 230 “protection”, seemed pretty troubling to the Democratic justices.
• We know that Justices Thomas, Alito, and Gorsuch have expressed sympathy for the arguments behind the social media regulation laws passed in Texas and Florida. They each asked questions about whether algorithms that were “neutral” should be legally distinct from those that were not. Justice Gorsuch further wondered whether any algorithms could be neutral?
• Not only does tomorrow appear to be a part-two of today’s argument in the minds of many of the justices, but it’s impossible to believe that are not thinking about how these internet liability cases relate to the state social media cases that are likely to come next.

FTC Commissioner Christine Wilson “Loudly” Resigns in Protest

Report from Bloomberg

In Brief – In the latest signal of dramatically deepening ideological and partisan divisions at the Federal Trade Commission (FTC), Commissioner Christine Wilson, announced her resignation in a scathing critique of the leadership of FTC Chair Lina Khan. The FTC is led by a board of five commissioners, three of whom are members of the President’s party, and two of the other party. Republican Noah Phillips resigned his seat in October and has not been replaced, leaving Wilson as the loan Republican. Wilson’s self-described “noisy exit” accused Khan, a high-profile progressive advocate of activist antitrust reforms, the Chair’s senior staff, and the other Democratic Commissioners, of operational and ethical failings, as well as promoting policies that Wilson claims exceed the legal authority of the FTC. The three Democrats released a statement saying that they “respect” Wilson for “her devotion to her beliefs” and “wish her well in her next endeavor.”

Context – Although Khan earned significant Republican support when the Senate confirmed her as an FTC Commissioner in 2021, that seems forever ago. Her tenure has been increasingly partisan and divisive. Senate Republicans challenged her in a September committee hearing and House Republicans are now planning aggressive oversight hearings. And Wilson made headlines last year when she accused Khan’s policy team of Marxist thinking. When the FTC Democrats released a policy statement in November outlining the FTC majority’s new legal thinking on “unfair methods of competition”, there was no bipartisan backing. Planned rulemaking on contentious issues like privacy (“commercial surveillance”) and Gig work, without Congress legislating, will exacerbate frictions. Finally, the Supreme Court’s revival of the “Major Questions” doctrine that limits agency authority to meaningfully change a federal standard absent legislative direction, as well as an upcoming High Court ruling on the constitutionality of the in-house FTC courts, will add to the conflict. It will be interesting to watch whether Republicans will allow any legislation that includes expanded FTC authority will move in either House of Congress.

Amazon Acknowledges Efforts to Address Regulator Concerns with iRobot Acquisition

Report from the Reuters

In Brief – Amazon has publicly acknowledged that it is engaged in talks with antitrust regulators to address concerns with its $1.7 billion acquisition of iRobot, the manufacturer of the Roomba robotic vacuum devices. The comments from an Amazon spokesperson followed a Financial Times report that the European Commission was actively moving toward a formal investigation of the deal. The focus of the Commission is said to be on privacy and data collection, in particular the type of data, potentially including sensitive in-home images, gathered by the robotic vacuums. Amazon claims that the devices only have basic room-mapping technology and a spokesperson said “We’re working cooperatively with the relevant regulators in their review of the merger” without confirming the FT report. Consistent with its stated desire to work with regulators, Amazon has not formally notified the European Commission of the acquisition, a step that would start a statutory timeline requiring the Commission to formally state its concerns, and kick off a more thorough review, or allow the transaction to proceed.

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, to less traditional charges like the privacy scenarios mentioned above. The US FTC, led by Lina Khan who came to prominence as a progressive antitrust reformer challenging Amazon’s business model, has formal investigations underway of Amazon’s iRobot and One Medical acquisitions, and it is reported they might become folded into a broad antitrust case targeting the ecommerce and logistics giant. There are also efforts in the UK pressing the Competition and Markets Authority to formally raise objections to the vacuum deal and create a third regulatory forum to undermine the transaction.

FTC Suspends (But Does Not Drop) In-House Challenge to Meta’s Within Acquisition

Report from Reuters

In Brief – Following its recent setback in Federal District Court, the Federal Trade Commission (FTC) has suspended its effort to block Meta’s $400 million acquisition of VR app developer Within Unlimited in the agency’s in-house court system. Last July, the FTC filed a federal lawsuit asking the court to prohibit Meta from completing the acquisition pending adjudication of the FTC’s legal challenge in its administrative court system. The two sides squared off in a week-long trial in December, and on January 31 Judge Edward Davila rejected the FTC’s bid to block the acquisition. While the FTC later announced that it would not appeal Davila’s ruling, the agency did not immediately confirm that it would drop its in-house court challenge, despite that being the norm in such cases. However, the agency has subsequently moved to pause its administrative court proceedings and cancelled the initial hearing, although it has not formally dropped the challenge and could move to restart the legal action.

Context – The Biden Administration’s antitrust enforcers have made it clear that they want to overturn court precedents and are willing to try to block deals where they know they might lose. So while the FTC failed to block Meta’s acquisition of Within, some see Judge Davila’s ruling holding some positive results for the progressive campaign to more effectively use antitrust laws to pare back the acquisition activity of big tech businesses. On the key question of market definition, the judge accepted the FTC argument that the relevant market was the narrow market of “VR dedicated fitness apps”, rather than broader markets for digital fitness services or VR apps in general, and that the market was “highly concentrated”. In addition, the judge accepted the FTC’s use of the doctrine of “actual potential competition” to show competitive harm. However, Davila said the FTC fell short by failing to make the case that Meta would have developed their own dedicated fitness app absent the acquisition.

Amazon’s Total Fees on the Average Third-Party Sale Exceeds 50% — So What?

Report from the Bloomberg

In Brief – Marketplace Pulse, the top research firm tracking online marketplaces, has released a report claiming that Amazon’s average cut of each sale made by a third-party seller surpassed 50% in 2022, the highest average Amazon fee since the firm began tracking the data in 2016. Amazon earns money on third-party sales through a variety of fees. Each sale includes a “referral” fee that ranges from 8 to 15 percent. But most third-party sales involve other paid services that dramatically increase Amazon’s overall commission. The two most impactful are fees paid to Amazon’s massive Fulfilment by Amazon (FBA) logistics system to store, pack, and deliver products, as well as “advertising” fees that a seller pays to bump their product up customer search results. Amazon’s average cut of third-party sales has been going up as more third-party sales involve FBA logistics and paid search preferences.

Context – Understanding how Amazon makes money in its ecommerce business provides valuable context on several public policy issues. First, while Jeff Bezos proudly claimed in 2018 that third-party sellers were “kicking our first-party butt” and outcompeting Amazon’s retail business, the truth then, and more so now, is that third-party sales using FBA are far more profitable for Amazon than direct retail. So, Amazon uses all its algorithmic levers to push FBA sellers into the “Buy Box”, a practice that is at the heart of the recently settled antitrust case in the EU, is under investigation in the UK, and should be the top issue if the FTC gets around to Amazon. The fact that most third-party goods sold on Amazon are housed in the company’s FBA system, a business unlike true marketplaces that generally do not handle sellers’ goods, is also a key issue on a range of product liability issues and challenges to Amazon from the FDA and CPSC. Finally, while Amazon’s 50% total fee on marketplace sellers seems high, Amazon-owned Twitch charges a 50% commission, and Amazon’s Kindle service charges commissions that reach 65 percent! Apple and Google must wonder about all the fuss over app stores charging 30 percent.

Apple Faces More Consumer Class Action Lawsuits Over Deceptive Data Practices

Report from Gizmodo

In Brief – Apple is facing a rapidly growing collection of consumer class action lawsuits following reports that the company engages in surreptitious and deceptive data collection, including when users have chosen Apple privacy settings that explicitly request not to have their data collected. Researchers claim to have demonstrated that extremely detailed and personal information is collected by a range of Apple’s apps, including the App Store, Apple Stocks, Apple Music, and Apple News, and sent to Apple linked to an ID number for a user’s iCloud account. The lawsuits contend that the company’s data collection practices run directly contrary to Apple’s very public pro-privacy claims and branding, the operating system’s settings that appear to give users the ability to opt-out of data tracking, and the highly controversial Apple policy changes in the past two years to allow users to reject ad-based data tracking by third-party apps.

Context – On one hand, Apple may end up facing the type of legal trouble that has plagued Google based on a one-time disconnect between its location tracking user settings on Android phones and Google’s actual data collection practices. In that case, an AP investigation in 2018 claimed that Google collected location data, in particular from 2014 to 2017, despite giving users options that appeared to allow them to opt-out. The company subsequently faced lawsuits in many markets. An Australian court sided with that country’s consumer protection regulator on similar charges in 2021. And in 2022 Google reached an $85 million settlement with Arizona’s AG, the first State AG to bring a complaint, as well as a $392 million settlement with a coalition of 40 other State AGs. Still outstanding are investigations by the European Union, as well as the Attorneys General of Texas, Washington State, Indiana, and Washington, DC, who have each brought separate state-level actions. On top of the consumer suits, one has to wonder if the company will face challenges from app developers who have attributed falling revenues in part to Apple’s controversial policy of allowing users to opt-out of ad-based tracking by third-party apps.

The Japan Fair Trade Commission Calls for Regulation of Apple-Google Mobile Duopoly

Report from the 9To5Mac

In Brief – The Japan Fair Trade Commission (JFTC) has completed its study of Apple and Google’s mobile-ecosystem duopoly — including app markets, operating systems, browsers, handsets, and wearables — and recommends that competition concerns be addressed through new ex-ante rules and regulations rather than traditional antitrust enforcement. The JFTC claims their market study of the two business models revealed strong network effects, user lock-in, and economies of scale, with limited competitive pressure from third parties or between the two giants. The agency calls for the companies to provide fairness and transparency in app store evaluation, ranking, terms of use; third-party in-app payments; fair use of data; and user data portability. In the device ecosystem, the report calls for lifting restrictions on handset makers to develop rival OS devices, a clear reference to Google’s handset policies, as well as opening operating systems to rival app vendors and browsers, a clear reference to Apple’s walled garden. The report acknowledges the privacy and security arguments the two companies use to justify a range of restrictions but does not propose how to address them in a future regulatory scheme.

Context – Japan, home to a very robust digital economy with a unique mix of domestic and foreign-based platform giants, has been a leader in the effort to improve the terms and conditions that digital commerce platforms provide to small business users, including through a balanced regulatory structure overseeing five giants – Amazon, Google, Apple, Rakuten and Yahoo Japan – including two based in Japan. If Japan pursues the mobile OS regulatory path outlined by the JFTC, a regime governing just US-based firms, it would appear to follow a model more like the Digital Markets Act in Europe. In the UK, facing similar mobule OS recommendations from the CMA, Apple asked how being forced to operate like Android would expand, rather than reduce, user choice. With similar regulatory ideas appearing to lose legislative momentum in the US, the prospect of different mobile ecosystems based on regulatory geography could become interesting.

EU Commission to Require Privacy Regulators to Regularly Report on Big GPDR Cases

Report from Bloomberg

In Brief – The European Commission has announced a major procedural change in how EU Member State data protection authorities report to the Commission, and the broader collection of other Member State data protection regulators, on the status of large-scale cross-border investigations they have underway based on the bloc’s landmark General Data Protection Regulation (GDPR). The GDPR, which assigns regulatory roles using the one stop shop (OSS) principle, has the data protection regulator in the country where a company has its EU headquarters play the key lead regulatory role. They will henceforth be required to submit a report every other month providing an update on the status of all major investigations involving companies impacting other Member States, including key procedural steps, investigatory measures taken, and relevant dates and deadlines. The new procedures are likely to apply primarily to actions involving large companies that operate across most or all of Europe.

Context – EU leaders are proud of the global uptake of GDPR-style privacy regulations. But privacy advocates and Big Tech critics are extremely frustrated that the GDPR has not resulted in a string of massive penalties being imposed on US digital giants. They often blame the OSS policy and criticize the privacy regulators in Ireland and Luxembourg, the two countries that are home to the European headquarters of most of the largest US-based digital firms (and TikTok too). Back in 2021, public criticism of the Irish Data Protection Commission by other Member State privacy regulators led EU Commissioner Vera Jourova to call for an end to the spats and to urge regulators to work cooperatively. A former EU commissioner from Luxembourg even recommended reducing the role of small state regulators. The GDPR’s OSS has been pared back in recent European High Court rulings and the EU’s two landmark digital platform legislative measures enacted last year conferred regulatory authority for very large digital platforms on the EU Commission rather than following the GDPR’s OSS model.

Google Giving Answers Not Links Is Not New – ChatBots and the One Box Problem

Report from the Washington Post

In Brief – On the heels of the public sensation caused by OpenAI’s release of ChatGPT, an AI-powered chatbot that generates responses to text inputs in a conversational manner, the potential for such tools to transform online search has driven several of the largest digital platforms, in particular Google, Microsoft, and Baidu, to publicize their work in the field. Microsoft, a top investor in OpenAI , claims to be incorporating the technology into its Bing search engine. Google, with dominant search market shares outside China, South Korea, and Russia, quickly reported on the state of its AI-powered conversational search tool but used an example that included a factual error and suffered a major negative hit in equity markets. The potential shift to online search engines delivering conversational “answers” in search results, rather than linking users to sites on the internet, is increasing concerns among online content creators that their work will be used by the AI-enabled machines to build answers without directing users to their sites.

Context – At PEI, we stick to public policy and don’t report general tech market news, but the Post article raises a long-time digital policy issue that you should note. The final third, starting with paragraph 13, shifts to a discussion of a couple of important realities in the evolution of general or “horizontal” internet search, in particular on Google, the dominant search engine. First, they have been using various forms of AI for a long time to better ascertain and respond to user queries. Remember, AI is a very amorphous concept. Second, Google has been providing “answers” rather than links, or preferencing them over links, for many years. Rather than presented as a type of chat, the “answer” is often presented in what Google calls a “One Box”. They have faced ongoing antitrust charges that they downplay links to specialized search sites and use One Box answers to preference their own services. Google says “answers” respond to user preferences. Google is likely to face strict regulatory oversight, at least in Europe, of how they bring chat-style answers into search. It’s less clear about others.

Apple Defends Mobile Payments Tech Practices from EU Antitrust Charges

Report from Reuters

In Brief – Apple will use a closed-door hearing on Tuesday, February 14 to try to convince EU antitrust regulators that its policy of blocking third-party mobile wallet providers from accessing key technology on the iPhone and other Apple devices that is used for in-store “tap and go” payments is not an anti-competitive abuse. The hearing is the latest step in an antitrust case begun in 2020 and follows an initial determination made by the European Commission Competition Authority last April that that Apple had abused its dominant position in the provision of mobile wallets on its devices by restricting payments providers such as PayPal from accessing Near Field Communications (NFC) capabilities. Apple has restricted access to NFC for payments since 2015 when the company launched Apple Pay. Apple claims that its policies did not restrict competition in digital payments, citing thousands of banks and other payments providers working through Apple Pay to serve customers, and argues that restricting access to certain technology capabilities better protects iPhone users from frauds than more open models.

Context – This antitrust case is about payments but is NOT part of the ever-growing fight over “in-app payments” hounding Apple and Google. Those fights are about the fees, often reaching 30%, that app developers pay when they sell digital products and services using apps distributed through their respective app stores. While Apple and Google have each proposed opening in-app payments to other payments services, they are offering a discount of just 4%. On that front, it seems inevitable that we will see a battle over public utility-style fee regulation. Spotify, a leader of the Apple fees fight, especially in Europe, also complains about discriminatory Apple practices in the music and digital downloads space. This investigation of NFC practices is more aligned with those concerns. Finally, its an open question how the emerging regulatory structure of the EU Digital Markets Act will operate in parallel with antitrust investigations involving the same giant companies.

European Parliament Expected to Deliver Approval of Its AI Act Next Month

Report from the Reuters

In Brief – The European Parliament aims to wrap up its legislative work on the AI Act next month, and backers hope the full EU legislative process can be concluded before the end of the year. The complex legislation establishing a regulatory framework to govern technology incorporating artificial intelligence (AI), including establishing risk tiers of AI uses, was proposed by the European Commission in 2021. The bill’s parliamentary leaders admit that the process has taken longer than expected, with the measure criticized by some for being too lax in addressing potential risks, while others criticize efforts to make the rules more strict as a threat to technology innovation. The recent popularity of AI-enabled chatbot tool ChatGPT, and reports of AI-powered online search tools from Google and Microsoft, have been used by AI regulation backers to increase momentum for the legislation. Following action by the European Commission and Parliament, the Council of Member States must still finalize its work on the measure so that the “trilogue” process to reach a final result can begin. During its six-month term leading the European Council, the Czech Republic worked to complete the Council’s AI Act but faced concerns from some Member States about the broad scope of the mandates.

Context – The fact that AI is such an amorphous, nebulous concept, as well as that relatively small firms have proven capable of big (and sometimes disconcerting) breakthroughs — see Clearview AI – adds complexity and uncertainty to the regulatory efforts. Happening alongside the EU’s AI legislative effort, the UK Government has proposed an AI Action Plan that eschews a centralized regulator or single legislative standard, and explicitly draws distinctions with the EU’s approach, claiming the bloc lacks flexibility and undermines innovation. The Biden Administration continues the US Government trend of sticking to high level principles.

President Biden Rallies Against Big Bad Tech in State of the Union (Does it Matter?)

Report from Bloomberg

In Brief – President Biden included two major legislative initiatives intended to restrain the business practices and influence of the largest digital platforms in his second State of the Union address. In a section of the speech on economic policy, he called on the Congress to enact antitrust reforms intended to prohibit digital giants including Amazon, Apple, Google, Meta, and Microsoft from preferencing their products and services on the platforms they operate. Later in the speech he called for privacy legislation to regulate the data practices of online companies, as well as returning to a topic that he raised in the speech last year, which is the regulation of social media platforms when used by kids and teenagers.

Context – If you read about Big Tech in President Biden’s SOTU and it’s described as a “win for progressives” or “an attempt to rally bipartisan support”, the writer has an agenda. On antitrust, Biden said “Pass bipartisan legislation to strengthen antitrust enforcement and prevent big online platforms from giving their own products an unfair advantage.” That one sentence was followed by 16 sentences on “junk fees”. Later on he returned to tech with “We must finally hold social media companies accountable for the experiment they are running on our children for profit. And it’s time to pass bipartisan legislation to stop Big Tech from collecting personal data on kids and teenagers online, ban targeted advertising to children, and impose stricter limits on the personal data these companies collect on all of us.” Two sentences. But then veterans programs got five. In short, 78 words in a 20-page, 1 hour and 13-minute speech. And nothing new. The White House claimed to lobby for those issues last Congress, when chances for success were higher. Last year’s SOTU even had the Facebook Whistleblower in the audience. It is noteworthy that Sec. 230, the key internet liability law, was not mentioned. The reality is that Biden Administration is backing an effort to have the Supreme Court rule that algorithmic recommendations and rankings are not covered by the law, a legal change that would be far more negative for platforms than anything Congress will do on internet policy this year.

CHouse Republicans Asking for Documents Detailing Feds Pressure on Social Media

Report from the The Hill

In Brief – The House Judiciary Committee leadership is formally requesting that the Department of Justice supply it with a set of documents that allegedly reveal numerous instances of federal government officials pressuring social media companies to ban or restrict user-generated content, possibly in violation of the First Amendment. The documents were produced as part of discovery in a federal lawsuit filed last May by the attorneys general of Louisiana and Missouri accusing a number of senior Biden Administration officials of pressuring Meta, YouTube, Twitter, and other social media platforms to suppress content that ran contrary to government policy on issues such as COVID, vaccines, and election integrity. The proceedings have already resulted in depositions of numerous administration officials and the disclosure of emails and other communications where government officials pressed company executives to exercise their authority more aggressively to moderate content.

Context – The House Government Oversight Committee hearing where Republicans and Democrats harangued four former Twitter officials and criticized each other over online censorship and misinformation, especially over Hunter Biden laptop stories, generated the most “social media censorship” news this week. But it’s not the most important story. Censorship of the Hunter Biden laptop affair (and other Republican charges from 2020) always had a major flow – Donald Trump was the President at the time. Anti-conservative government censorship via coercion of, or delegation to, the platforms never made much sense. It requires a kind of rogue government or “deep state” conspiracy theory. But actions by Biden officials in 2021 to push platforms to better police “misinformation” on COVID and vaccines is a more thoughtful delegated-censorship theory and Republicans digging there is a more reasonable exercise. A related bill from House Republican leaders to prohibit federal officials from pressuring social media platforms earned “kudos” from digital policy expert MikeMasnick who said the bill is “focused on actually reinforcing the 1st Amendment’s protections.”

Microsoft Gets EU Statement of Objections on Activision-Blizzard Acquisition

Report from Bloomberg

In Brief – The European Commission has sent its formal list of competition concerns to Microsoft related to its $69 billion deal to buy giant video game maker Activision Blizzard. The statement of objections outlines the issues that Microsoft must address to the satisfaction of EU regulators to avoid the deal being rejected. When the Commission announced in November that it was pursuing a more thorough examination of the acquisition, it outlined concerns that Microsoft might use access to top Activision games to distort competition in game consoles, subscription services, and cloud-based gaming, and design games to work better on the Windows’ operating system to disadvantage rival operating system providers. A company spokesperson said Microsoft is “continuing to work with the European Commission to address any marketplace concerns.”

Context – Microsoft’s Activision deal, the biggest digital acquisition of all time, is providing insights into the current phase of global merger reviews where increasingly activist regulators, and companies themselves, forum shop to get their best result. In the US, the FTC filed suit in its in-house administrative court, bypassing going to US District Court to get an injunction (the forum where it failed to block Meta’s bid for VR startup Within) because it felt it could rely on delays caused by the EU and UK. The UK Competition and Markets Authority, has delivered its provisional findings confirming objections to the deal, focusing on on consoles and nascent cloud gaming, with a deadline of April 26 for their final Phase 2 decision. Concessions to reach agreement are still possible. The CMA is seen by some analysts as the key hurdle to the deal due to the deference given to the regulator by the UK’s special competition court. Microsoft has been engaged in a , including acquiescence to labor organizing, support for tech antitrust reforms, agreeing to software license changes to promote European cloud services competition, and support for app store regulation. If Microsoft addresses EU and UK concerns with an acceptable set of concessions, they hope they can be used to wrap up the deal in the US as well.

Report from the Washington Post

In Brief – After months of speculation that the Committee on Foreign Investment in the United States (CFIUS), an opaque multi-agency panel that reviews acquisitions of US firms by foreign entities, might formally review Elon Musk’s acquisition of Twitter, the panel reportedly does not believe it has legal jurisdiction to scrutinize the deal. CFIUS has authority to examine acquisitions of US companies by foreign owners. Musk, an American citizen, acquired Twitter, then a public company, for $44 billion, taking it private. The billionaire financed most of the purchase on his own, but $2.5 billion was reported to be from foreign entities, including ones linked to Saudi Arabia and China. Democratic policymakers, including President Biden, had publicly speculated that Musk and the deal should face a security review.

Context – The prospect of a major social media platform, especially one central to the lives of the media and academia, being led by an aggressive critic of progressive worldviews who promised to change content moderation policy, was guaranteed to set off alarm bells on the left and generate applause on the right. Progressives quickly called for government interventions, including on antitrust grounds, which never was likely, and on security questions. The lack of concrete review standards for CFIUS made that a bigger risk, but that progressive Hail Mary pass now appears to have failed as well. There are real and important digital platform issues where Musk owning Twitter may shake things up. The fact that it is not inevitable that every big social media platform will be run by ideological progressives forever may change how some Republicans think about regulation of online content moderation. On Capitol Hill, it may push Republicans and Democrats even further apart, and the House Government Oversight Committee hearing on past Twitter content practices will be an early example. A more important question is whether the new reality will impact the thinking of the conservative Supreme Court justices who have called for reevaluating constitutional limits on government regulating social media content moderation?

EC Kicking Off Consultation on Forcing Big Digital Platforms to Pay More to Telecom Giants

Report from Euractiv

In Brief – After months of talk from top EU officials indicating sympathy for the campaign by Europe’s largest telecommunications companies to receive new payments from the largest digital platforms, the European Commission will circulate a questionnaire to digital platforms, network operators, and other internet industry stakeholders, as part of a consultation on the issue. The idea is often described as “sender pays” or the “fair share” proposal. The lengthy Commission questionnaire is reported to query a range of issues including how to determine whether, and if so, how, large digital platforms like Google, Meta, Amazon, and Netflix should be required to provide new payments to large consumer-facing telecommunications companies like Deutsche Telekom, Telefónica, Orange, and Telecom Italia. Along with querying the investment contributions and plans of all major market participants, the Commission is asking about the impact on consumers, the open internet, net neutrality, and the safety and robustness of the underlying network.

Context – Telecommunications companies have accused the largest internet platforms of “free riding” from the early days of the commercial internet. This is despite every internet user, from smallest to biggest, paying for the bandwidth they use. These arguments have generally been part of net neutrality debates because telecom companies advocating for corporate “senders” paying more have proposed offering some manner of better service in return. But that means those who don’t pay have unequal access to users. The telecom industry champions of the current campaign propose no benefits to the payers, which therefore seems more like a tax, but they claim it’s not a tax. The battle lines are well established. European consumer advocates, smaller internet network infrastructure firms, and industry regulators have expressed opposition to aspects of the fair share plan. But the expected payers are largely US-based, and the expected recipients are European-based, so that is a political asset.

Amazon Faces Dark Patterns Probe in Poland Over Product Delivery Terms

Report from the TechCrunch

In Brief – Poland’s national competition and consumer watchdog, the UOKiK, has opened an investigation of Amazon for misleading consumers over delivery times, product availability, and the sales contract that applies when products are purchased on the site. The ecommerce and logistics giant is accused of employing misleading practices that often fall under the broad descriptor of “dark patterns” to deceive consumers regarding the availability and delivery of products as they are making buying decision that might not be made if consumers understood the actual availability and expected delivery of the product. For example, Amazon presents a shopper brightly colored “Buy now” and “Proceed to checkout” buttons to nudge consumers to place an order based on prominent availability and delivery dates that Amazon does not actually stand behind, while at the bottom of the website page in small print using less a visible color, are terms of sale stipulating that the sales contract is not actually concluded until the product ships, which may be delayed far beyond the terms presented.

Context – Amazon was accused of deceptive dark patterns related to its arduous unsubscribe process for Prime subscribers in 2021 and agreed last year to dramatically simplify that website flow in Europe. The EU’s Digital Services Act includes a high-level prohibition on dark patterns but focuses on three practices – fake countdown timers, the use of design, colors, or language to nudge users to a preferred option, and hiding information on a product or service by making it less visible on a website page. A coalition of European consumer protections authorities recently conducted a sweep of 399 sites focused on those practices, finding fault with 148. In the US, congressional legislation has been introduced to regulate a range of practices, and the FTC has investigated the issue, but don’t expect meaningful federal action anytime soon. A few states with online privacy laws, including California, nullify user consent when achieved using deceptive practices, which may prove impactful if state regulators aggressively enforce the laws.

Spanish Judge Rules that Amazon Gig-Work Delivery Drivers are Employees

Report from Bloomberg

In Brief – A Spanish court in Madrid has ruled that people who delivered packages for Amazon through the company’s Gig-work Flex program in Spain should have been classified as employees not freelance workers, rejecting Amazon’s argument that the company was an intermediary between independent contractor drivers and retailers selling goods through its online marketplace. Amazon’s Flex delivery program, which operates with an Uber-like app and pays individual drivers to deliver packages in their own vehicles, was begun in the United States in 2015 and expanded to Spain three years later. Amazon ended the program in Spain in 2021 when the country enacted legislation requiring large Gig style ridesharing and delivery platforms to reclassify drivers as employees and extend a range of labor rights, benefits, and protections. Amazon is awaiting a decision in a similar case in Barcelona.

Context – Amazon is one of the largest Gig work businesses. Two Gig platforms dominate Amazon’s delivery services. Uber-like Flex is one. The other is the Delivery Service Partner program, which is a fleet of nominally independent small businesses who buy Amazon-branded delivery vans, often using Amazon credit, and often use Gig drivers to drive them. Driver performance is tightly guided by Amazon. When safety, health, or labor shortcomings occur, Amazon pins the failings on the small firms. The European Commission has proposed a directive to regulate Digital Labor Platforms in Europe, and tackle “bogus self-employment” and “algorithmic control” of workers. The European Parliament has concluded its work and voted to proceed to final “trilogue” negotiations with the Commission and European Council of Member States. Employment regulation is an area with meaningful differences in national laws between EU member states, including on the regulation of Gig work platforms, and the 376-212 vote in Parliament reflected that lack of consensus. The European Council of Member States must still arrive at a collective position before wrap-up negotiations can begin.

Biden NTIA Supports App Store Regulation in Run Up to Tech Reform in State of the Union

Report from the Yahoo Finance

In Brief – The Commerce Department’s National Telecommunications and Information Administration (NTIA) has released a report describing the mobile app ecosystem and app store practices of Apple and Google as harmful to consumers and app developers. The report calls for the companies to make changes, but also backs federal legislation and enforcement actions. NTIA is not an enforcement agency and regulatory action would require the FTC or the Department of Justice.

Context – The Biden Administration likes publicizing its Big Tech regulation agenda. The President is expected to take aim at Big Tech in his State of the Union (SOTU) address, including antitrust, privacy, or social media legislation. Last year’s SOTU praised the Facebook Whistleblower and called for legislation regulating how websites handle teen users. Last September the White House hosted a tech policy event to energize their progressive supporters, releasing a report calling for action on antitrust, privacy, Sec. 230, and regulation of social media algorithms and content moderation. They were public backers of tech bills during the end-of-the-year Lame Duck session. All fell short. Are the new legislative calls more political or more substantive? Political. Most of what’s changed from last year to this year reduces prospects for major bipartisan tech legislation in Congress. The House Republican Leadership is very wary of the progressive antitrust agenda and Biden’s activist regulators. They put a skeptic in charge of the House Antitrust Subcommittee, passing over a top Republican champion of Big Tech antitrust reform. The Republican focus will be on scrutinizing social media content moderation for bias and pressure from Biden Administration officials. For impactful action, look to the federal courts, especially the major Sec. 230 cases in front of the Supreme Court later in February, and the activist regulatory agenda at the FTC. The biggest recent Biden action is not talk backing legislation, but their amicus brief calling for the Court to strip Sec. 230 protection when online content is ranked or recommended by algorithms, meaning almost everything.

Department of Justice Scrutinizing Child Welfare Algorithm for Discrimination

Report from AP

In Brief – Civil rights attorneys of the US Justice Department (DoJ) are scrutinizing a controversial data analysis tool used by a Pittsburgh-area child protective services agency to help identify families and children for follow-up, including further investigation. The Allegheny Family Screening Tool, used since 2016, employs algorithms to analyze a wide range of data points, and has been criticized for being biased against people with disabilities, including families with mental health issues. Child protective services workers are often accused of over-interventions in some cases, and not giving enough support to at-risk children in others. Data analysis tools are backed by supporters as ways to improve the efficiency of child welfare officials and better help identify children at greatest risk. Critics argue that predictive algorithms can reinforce biases based on race, income, disabilities, or other external characteristics.

Context – This story of “a controversial artificial intelligence tool” is a great example of the shadowy line, if any line exists at all, between long-practiced data analytics and AI. Super-sharp policy analyst Jim Harper makes the case that using the term AI causes more harm than good for policy thinkers. “Algorithms” are just computer code. Digital tools can be flawed and can knowingly or unknowingly skew results, including in discriminatory ways. But that’s been the case for many years. The child welfare tool in this case was introduced in 2016. A recent report criticizes a long-used IRS data analytics tool that recommends which taxpayers to audit for leading to over-auditing of Black Americans because the tool targets recipients of the Earned Income Tax Credit, an income support program, and Black Americans are disproportionately lower income. That’s not flawed AI, it’s flawed IRS policy. Finally, I can’t mention AI without mentioning Open AI’s ChatGPT and fear in the education establishment (which seems like Wikipedia fear two decades ago). So now there are AI text detectors. OpenAI’s tool can be accessed here. I had it review a recent PEI email. It said “very unlikely” written by AI. I feel a bit disappointed.

TikTok’s CEO Scheduled to Appear Before US House Committee in March

Report from the Wall Street Journal

In Brief – TikTok’s CEO is scheduled to testify before the House Energy & Commerce Committee on March 23. It will be the first time the top executive from the Chinese-owned short-video social media phenom will appear before Congress. Top issues likely to be raised include company data and privacy practices, including past abuses related to users under age 13, the impacts of a wide range of objectionable content on children and teens, and concerns over the potential influence of the Chinese Government on TikTok and its parent ByteDance, both by covertly pushing the platform to restrict or promote sensitive topics to US users, or by accessing user data that could be used in influence operations.

Context – Befitting its ascension into the stratosphere of digital giants, TikTok is embroiled in a range of legal and regulatory challenges in many markets. Some are basically the same issues facing the other social media giants. These include privacy, especially the handling and use of data from children, and the potentially negative effects of objectionable content, especially on teens. On those issues, TikTok is in the same group as Meta’s Facebook and Instagram, YouTube, Snap and Twitter. However, the fact that TikTok and parent ByteDance are Chinese companies makes a real difference on other issues. Many security officials genuinely doubt that the influence of the Chinese Government, which is heavily invested in digital surveillance and control over online messages globally, can be limited. This is the real risk to TikTok. The company is reported to be ready to publicly defend a plan it has developed to use Oracle, and some form of semi-independent corporate security group overseen by US security officials, in intermediary roles to wall off US operations. But without believing that Chinese Government influence over tech companies can be limited, and events like the Jack Ma saga reinforce that it can’t, any proposal has problems. At the same time, banning the app, like India has done, likely faces big legal hurdles. And we saw in 2020, with the TikTok and WeChat bans, that the US judiciary is an independent branch that can check government actions.

Effort to Organize Programmers at Boston Game Studio Suffers Setback

Report from ArsTechnica

In Brief – Following reports from union organizers that a majority of workers at Boston-based video game studio Proletariat had submitted union cards indicating support for creating a union, the Communications Workers of America (CWA) called off their union drive. The labor organizing group blamed the turnabout on the studio’s top executive and founder meeting with employees to argue against the organizing drive. Other employees took to social media to defend the meetings and criticize the labor organizers for misrepresenting the level of support for the labor drive among the employees. Proletariat, which was recently acquired by giant game developer Activision Blizzard, is a relatively small studio, with approximately 100 employees. After the CWA announced what they claimed was strong support for the Proletariat Workers Alliance, which extended across the full workforce including game designers, animators, engineers, producers and quality assurance workers, the studio management rejected the call to recognize the union and said they would ask the NLRB to oversee a vote.

Context – The video game industry has recently been at the forefront of the highly politicized effort to organize digital workers. However, recent successes at two Activision Blizzard studios and Microsoft-owned Zenimax involved only quality assurance “game testers” who are generally hourly employees earning around $20 per hour, not programmers or developers. Those unions are like efforts at Amazon fulfilment centers and Apple retail stores. Tech companies but not tech labor. The Proletariat union, while small, claimed to include programmers and other skilled workers, so the setback reinforces the labor status quo. At French game studio Ubisoft, some programmers reacted angrily to a recent CEO exhortation for the staff to redouble efforts in tough economic times, leading to a small half-day walkout. And the non-traditional Alphabet Workers Union has accused the company’s YouTube office in Austin, Texas of using return to work plans to derail efforts to organize non-employee contract workers.

Federal Judge Rejects FTC Effort to Block Meta’s Acquisition of VR App Developer

Report from the Wall Street Journal

In Brief – In a major setback to the Federal Trade Commission’s Democratic majority led by Chair Lina Khan, a high profile progressive antitrust reform advocate, Federal Judge Edward Davila rejected the agency’s bid to block Meta Platform’s acquisition of VR app developer Within Unlimited for $400 million. In July, the FTC’s three Democratic commissioners filed a federal lawsuit to delay the acquisition and allow it to challenge the deal in the agency’s in-house administrative court. When the two sides squared off in federal court in December, the FTC argued that Meta acquiring Within, the developer of a leading VR fitness app, would undermine VR competition in the future because Meta would otherwise be forced to develop an app that competes with Within’s product. Meta argued that the FTC suit was “based on ideology and speculation, not evidence” and would send “a chilling message to anyone who wishes to innovate in VR.” CEO Mark Zuckerberg argued that his company was helping to build a nascent VR industry for a wide range of developers and companies, and that while fitness apps could prove popular to some users, Meta was focused on a wide range of applications.

Context – Although the top competition regulators in the US, EU and UK all express deep concerns with Big Tech acquisitions, there is no prevailing theory behind a change in merger reviews. Antitrust enforcers have appeared most willing to challenge Big Tech deals to buy startups. The UK CMA unwound Meta’s deal for Giphy, the FTC tried to block Meta’s bid for Within, and the EU is fighting the Illumina-Grail deal. But some critics have called out the activist enforcers for shying away from the big legal battles. Microsoft’s Activision Blizzard deal, the most expensive digital acquisition ever, offered regulators the chance to go big. The FTC has sued to block that deal, and the European Commission is expected to file objections as well. The Biden Administration’s antitrust enforcers admit to wanting to change court precedents by trying to block deals where they know they might lose. So maybe they see this Meta setback as a win of sorts.

European Commission Readying Antitrust Investigation of Microsoft Teams Service

Report from Politico

In Brief – The European Commission Competition Authority is reported to be planning to open a formal antitrust investigation into Microsoft practices regarding its video and messaging service Teams. The investigation is the result of 2020 antitrust complaint from Slack, a leading chat-style, remote collaboration platform, alleging that Microsoft unfairly bundles its Teams product with its market-leading Office 365 software package. In preparing to open the formal review, the Commission has informed Microsoft Teams rivals and customers of some of the data and evidence they hold that the Commission plans to use in making its case. Microsoft has faced similar complaints regarding tying and bundling from European cloud services providers, who have argued that the software giant uses terms and pricing of its popular Windows 10 & 11 software and MS Office Suite to promote sales of its Azure cloud service by dissuading software customers from choosing alternative cloud vendors. In response to those charges, Microsoft has rolled out a plan to change license terms to reduce the cost for EU-based cloud companies to host top Microsoft software on their cloud infrastructure.

Context – Despite its standing as a digital giant in software, cloud services, gaming and business services, Microsoft largely avoided the first years of the antitrust “techlash”. However, the company’s acquisition of video game giant Activision Blizzard, which is the most expensive acquisition ever by one of the digital giants, has sparked scrutiny from US, EU, and UK competition enforcers. Besides engaging in a full-throated defense of the acquisition as good for consumers and developers, Microsoft has also pursued a global goodwill campaign aimed at progressive regulators. It has included acquiescence to unprecedented tech labor organizing, support for major antitrust reforms, vocal support for app store regulation aimed at Apple in particular, and the aforementioned software license changes to promote European cloud services competition. The European Commission may be looking to add greater competition for Teams during the pending Activision review as well.

Google Implements Android Changes in India in Response to Antitrust Rulings

Report from the Reuters

In Brief – Google has announced policy changes to comply with the remedies ordered by the Competition Commission of India (CCI) following their determination that the company illegally leveraged its dominant position in markets such as online search and Android app stores to protect the market position of its apps. Android users in India will be able to choose their default search engine when they first set up an Android device, as well as uninstall the first-party apps that come with their device. In response to device maker concerns, Google says it will update the Android compatibility requirements to allow partners to build “forked” variants. Finally, Google will allow third-party billing options for in-app purchases within six months. The changes were set in motion following the Indian Supreme Court’s rejection of Google’s motion to stay the CCI’s orders while the company appealed the agency’s decision.

Context – Both Apple and Google are under fire for allegedly anticompetitive practices within their competing mobile ecosystems. But they follow very different models. Apple has long had a “walled garden” business model where they make all the devices, refuse to license the operating system, and strictly control apps. Google’s Android ecosystem is a very different hybrid model. Android source code can be used in an “open source” manner, but manufacturers who produce devices they desire to brand as official “Android” devices (think green robot logo) must follow many contractual mandates, including regarding apps, for all their Android-compatible 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. The South Korean Fair Trade Commission, responding to Samsung and LG concerns, has been pressing Google to open up on allowing alternative “forks”, as well as taking up the cause of alternative in-app payments, which many app developers hope will result in lower fees overall.