Rep. Jim Jordan (R-Ohio) looks on during a media availability after he was elected to be House Republican Leader for the upcoming 120th Congress on Capitol Hill on Nov. 10, 2026. (Francis Chung/POLITICO via AP Images)

Will a New Congress Ease the Pressure on Europe’s Digital Regulations?

Campaigners march against data centers in Dublin, Ireland, on November 6, 2026. Photo by Paula Geraghty.

Ireland is Helping Shape a Data Center Boom That Could Benefit Big Tech

Members of the Garda Armed Support Unit outside the National Gallery in Dublin as more than 50 world leaders will visit Dublin for the European Political Community (EPC) summit on Thursday, followed by an informal meeting of the EU Council on Friday. The biggest security operation in the history of the Irish state is under way, as authorities in Dublin mount a ring of steel throughout the capital. Picture date: Wednesday November 11, 2026. PA Photo. Photo credit should read: Niall Carson/Press Association IRISH (Press Association via AP Images)

Europe’s AI Future and Security Challenges Take Center Stage at Dublin Summit

OpenAI CEO Sam Altman delivers the keynote address during the OpenAI developers conference at Fort Mason on September 29, 2026 in San Francisco, California. (Photo by Heather Diehl/Getty Images)

Friends of the Court Flock to the OpenAI Copyright Fight

Sam Altman watches over a family consuming his AI. The living room vignette is framed by power tool imagery. The words “Behaviour Power” are the top of the artwork, framing the intent of the piece.

Behaviour Power II by Bart Fish & Power Tools of AI / 

Humans Have Rights. Machines Do Not.

A blue coloured icon of a data server, surrounded with comic-book style words 'Gasp!', 'Boom!' and 'Pow!' in colours of yellow and red.

Don't Let AI Become the Internet's Gatekeeper

The United Nations headquarters in New York. Shutterstock

The United Nations headquarters in New York. 

A Defining Moment for the United Nations and the Future of AI

An illustration of Tigrayan university chemistry Professor Meareg Amare by Alina Pechenckina. Source: Amnesty International / CC by 4.0

An illustration of Tigrayan university chemistry Professor Meareg Amare by Alina Pechenckina. Source: 

Meta Returns to Kenya’s High Court Over Hate Speech in Ethiopia

Executive Vice-President for Tech Sovereignty, Security and Democracy of the European Commission Henna Virkkunen, and European Commissioner for Democracy, Justice, the Rule of Law and Consumer Protection Michael McGrath speak at a press conference on Wednesday. Source

Executive Vice-President for Tech Sovereignty, Security and Democracy of the European Commission, Henna Virkkunen, and European Commissioner for Democracy, Justice, the Rule of Law and Consumer Protection, Michael McGrath, speak at a press conference on Wednesday. 

Assessing the EU Digital Fairness Act and Its Promise of Consumer Protection

A Close Read of Big TikTok and Meta Settlements Raises Important Questions

A collage of circuit boards interwoven with aerial photographs of water treatment facilities, agricultural lands, industrial sites and mining machinery.

Brazil's AI Development Strategy Needs an Edge Layer

Anthropic's logo is displayed on a smartphone with a stock market chart in the background on October 27, 2026, as Anthropic filed a Form S-1 with the Securities and Exchange Commission (SEC) for a planned initial public offering in the United States. (Photo by Samuel Boivin/NurPhoto via AP)

Could the United States Seize Anthropic at a Discount?

Journalists test AI software used to manipulate images at Phnom Penh Municipal Police headquarters following an operation against scam centers in Phnom Penh, Cambodia, Tuesday, Sept. 22, 2026. (AP Photo/Heng Sinith)

Cambodia’s Scam Crackdown Has Failed to Stop the Industry’s Global Spread

Two women sitting on a bench smiling with a US flag and an 'I voted' button

Photo by Pocstock, an inclusive, BIPOC-owned media and stock photo company that features real, human-captured photography.

Tech Bros Are White, Progressives Afraid, But Black Women Public Interest Tech Leaders Are Brave

The TikTok logo is displayed through a magnifying glass in this photo illustration in Ontario, Canada, on February 5, 2026. (Photo by Thomas Fuller/NurPhoto via AP)

Can the EU Make TikTok Take Responsibility for Its Rabbit Holes?

In the foreground, multiple people are surrounded by computer screens filled with data visualizations, charts, and technical information. A woman seated in an armchair appears deep in thought, surrounded by data-filled monitors. Beside her, a man leans over, using a tablet to assist with their inspection of a plant or tree. In the centre, a figure holds a large frame or screen displaying anatomical illustrations, representing the use of AI to analyse medical imagery. To the left, another person is intently observing a computer screen, while a second figure nearby is deeply immersed in analysing data. A woman dominates the right side of the composition, gazing upwards as if in contemplation or envisioning something beyond the immediate scene. The background features more people, including a family holding hands, and other abstract representations of data.

AI is Everywhere by Ariyana Ahmad & The Bigger Picture / 

In the AI Era, Language is Now an Attack Surface

Close up photo of a pile of US 25-cent coins

10 Things States Should Do with the Meta Settlement Money

The sun rises over the US Capitol and the Washington Monument on the National Mall the day after election day on November 04, 2026 in Washington, DC. (Photo by Kevin Dietsch/Getty Images)

A New Agenda for Tech Regulation Beyond AI

People attend a watch party for Georgia Gov.-elect Keisha Lance Bottoms at the Atlanta Marriott Marquis on November 3, 2026 in Atlanta, Georgia. Bottoms defeated Republican gubernatorial candidate Rick Jackson to become the nation's first Black woman governor. (Photo by Megan Varner/Getty Images)

US Midterms Results Present 'Real Moment for Change'

A woman wears a mask with a Wi-Fi symbol as a protest against internet censorship in Iran. Several hundred people demonstrated in Frankfurt, Germany on Rathenauplatz on May 16, 2026, demanding an end to the Islamic regime in Iran on placards and in chants. (Photo by Boris Roessler/picture-alliance/dpa/AP Images)

Freedom House Charts 'Seismic Shifts' on the Global Internet

Detail from Behaviour Power II by Bart Fish & Power Tools of AI / 

 a federal court in Florida to dismiss a wrongful-death suit, brought by the mother of a fourteen-year-old boy who had killed himself after months of conversation with its chatbot. The company made two arguments. One, that the chatbot’s words were speech that its users had a First Amendment right to receive. And two, that Character.AI was a service rather than a product — so product-liability law could not touch it. In May 2025, Judge Anne Conway 

Character.AI’s maneuver is part of a larger pattern. When the parents of a sixteen-year-old sued OpenAI over his suicide, the company’s 

 also pleaded the First Amendment: the claims would restrain protected speech. The very same document argues ChatGPT is offered “as is,” with no promise of accuracy, and that the boy had circumvented its guardrails. Elon Musk’s xAI (since renamed SpaceXAI) 

 a federal court in Colorado this spring that every choice in building Grok is an expressive act, and that Grok’s outputs are protected speech. When Grok praised Hitler last summer, xAI 

 the posts came from a code update independent of the model, and the apology 

 under Grok’s own name. Google, which spent two decades telling American courts that its search results are 

 over an AI Overview by saying the system merely summarized the web. It also 

 a Munich court in May that no one should trust an AI overview without checking it. Back in 2024, Air Canada 

 that its customer-service chatbot was a separate legal entity, responsible for its own actions.

When the output causes harm, the companies deny responsibility. It is the machine’s doing, or a product no sensible person would rely on, or the user’s misuse. When a state or a plaintiff would hold the words to account, the same output becomes speech, and the First Amendment applies.

Social media companies deployed an earlier incarnation of this strategy. They called themselves publishers when editorial discretion was profitable, and neutral carriers when faced with liability for what they published. The result was a privatized public square in which tech platforms were unaccountable. AI companies have taken this even further. A social media platform could at least point to a human author somewhere in the chain. A chatbot’s prose differs; no person at the company wrote it, read it, or chose it before it reached the user. Yet the companies ask that it be protected as though a human had written it.

The question grows more urgent as these systems stop talking and start acting. In July, a poorly controlled experiment at OpenAI resulted in its AI model 

 the network of HuggingFace. Since then both Anthropic and OpenAI have disclosed numerous instances of AI agents hacking outside networks in pursuit of their prompted goals. These are examples of AIs exhibiting 

: behaving in ways its controllers neither expected nor wanted. Nothing in the actions count as speech. Breaking into a server is an act that no person at OpenAI explicitly chose. Jill Lepore, 

, puts the blame on Congress, which has passed no federal statute governing AI agents.

The First Amendment protects expressive choice. It presumes a speaker who can be persuaded, shamed, jailed, or moved to change their mind. Its remedy for bad speech is more speech, which assumes that speech is costly and attention plentiful. A system that can produce millions of individually tailored political messages an hour reverses both assumptions. Verbiage that no person chose is a product, and the company that built the model should answer for it.

The industry wants a self-serving third category that bridges AI speech and action: AI output is human speech when expressive rights are at issue, and the machine’s own doing when blame is. The result is agency without accountability, something no human speaker has ever possessed. That’s why Massachusetts convicted a young woman of 

 in 2017 for urging her boyfriend, by text and by phone, to kill himself.

The Court has already glimpsed the problem. Upholding the TikTok divestiture law in January 2025 without dissent, the justices 

 that “new technologies with transformative capabilities” counseled caution, then decided the case on national-security grounds and 

It had postponed another decision six months earlier. In 

, the Court said that a platform’s curation is protected when it carries out the platform’s own editorial standards and set aside the case of an algorithm that responds only to user preferences. Justice Amy Coney Barrett, concurring, asked whether a human being with First Amendment rights has made any expressive choice when a platform’s owners hand the reins to an AI tool. The Court’s own doctrine already turns on the difference between the words a person chooses and the outputs a machine generates.

Congress can do more. It can enshrine the critical difference between human expression and algorithmic manipulation into law. The rule is simple: companies that build and deploy a machine must answer for what it does.

Legislating this has three steps. The first is to assign liability, so that companies that build or deploy an automated system are liable for its outputs — unless a human adopts the output as their own. In May, a Munich court provisionally ordered Google to stop serving an AI Overview that falsely accused two publishers of running scams, holding Google answerable for its summaries and refusing to excuse it as the mere host of someone else’s content. Companies cannot both be a conduit when sued and a speaker when regulated.

Second, the law should deem any optimized amplification, such as curated social-media feeds, as the conduct of the platform. This is different from the editorial judgments the 

 decision protects; ranking content by predicted engagement is not an editorial standard.

Third, the law should require disclosure for machine-generated political communication. This transparency is essential, so citizens know when no person chose the words addressed to them. Disclosure is the one remedy the Court has 

, even while striking down limits on political spending.

None of this would silence anyone. A person who uses a machine retains every protection they had before. Nor would it stop any output from reaching the listeners whose rights Character.AI invoked. Liability makes products safer, and disclosure tells listeners what they are hearing. Each provision puts a human decision behind the machine.

Of course, any law like this would draw constitutional challenges. That is the Court’s opportunity to rule on what it deferred in the TikTok and 

 cases: does the First Amendment protect anything that resembles speech, or only the people who choose to speak and to listen? Avoiding this question in the AI age will lead to a public square filled by machines that answer to no one. The alternative is a rule the Court can adopt without disturbing a single human speaker. Where no person chose the words, no speech right attaches.

There should be no agency without accountability. A machine may reason about human values, but responsibility for what it does stays with the people who engineered it. The companies have claimed First Amendment rights when regulation threatens and a bystander’s innocence when harm is done. Congress and the courts can foreclose this hypocrisy. Rights belong to human beings; the machines we build, however clever, are ours to answer for.

On Nov. 12, the case brought by two Ethiopian nationals, Abrham Meareg and Fisseha Tekle, and the Katiba Institute against Meta 

. The petitioners allege that Meta’s algorithms amplified hateful content in Ethiopia during the Tigray war between 2020 and 2022. Abrham Meareg’s father, Bahir Dar University chemistry professor Meareg Amare, was 

 after a viral Facebook post identified him as Tigrayan and alleged that he had links with the Tigray People’s Liberation Front (TPLF).

Facebook users called Meareg a ‘snake’, claiming that he posed a threat to the Amhara community. Some commenters complained that the Facebook posts served to warn Meareg rather than to 

, and urged readers to act with haste. On Nov. 3, 2021, three weeks after the first inciting Facebook post made against him, Meareg was followed home from his university by a group of men and 

. Tekle, an Amnesty International legal adviser, was 

 of being a TPLF agent and linked to a founder of the party. After the posts went viral, he says he 

 the case from being heard in Kenya, arguing that because the company is not registered in the country, it cannot be subject to Kenyan law. Meta also invoked the doctrine of constitutional avoidance, arguing that the dispute was contractual, arising from Facebook’s terms of service, and should therefore have been brought before a United States court, 

. However, the High Court refused to strike down the petition, holding that Facebook’s operation of a content review center in Nairobi when the petition was filed gave Kenyan courts jurisdiction. (Meta told the court that Sama stopped the work in March 2023, three months after Meareg, Tekle and the Katiba Institute filed the petition on Dec. 14, 2022.)

The court further held that the matters raised were substantial, transcended the interests of the parties, and raised questions about the protection of fundamental rights and freedoms in the digital era, and so met the threshold for certification of a bench under Article 165(3)(b) of the Kenyan Constitution. 

 of the Kenyan Constitution allows the Kenya High Court jurisdiction to determine the question whether a right or fundamental freedom in the Bill of Rights has been denied, violated, infringed or threatened. On Nov. 12, the Court will listen to Meta’s arguments opposing a recent application by Meareg, Tekle and the Katiba Institute seeking to add new evidence to their petition.

The case returns to court as the fragile Tigray truce established by the November 2022 Pretoria Agreement 

. In Ethiopia, where decades of repression have stifled media freedom, people turn to Facebook for information. Meta itself is aware of this. The Meta-funded Oversight Board, in a 

 to the company published in a case decision in December 2021, cited expert briefings noting that “Facebook is an important, influential and popular online medium for communication in Ethiopia” and that “there is little to no coverage on the conflict-affected areas in Ethiopian media, and Ethiopians use Facebook to share and receive information about the conflict.”

Nearly a year before war broke out in November 2020, David Kaye, then the UN Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression, 

 to establish regular contact with Ethiopian authorities and civil society groups to better understand conditions in the country and establish rapid-reaction channels to enable civil society to report concerning content. However, Facebook prioritized the use of ‘

’ over outright removal of hateful and inciting posts to avoid negatively impacting the corporation’s Meaningful Social Interaction (MSI) metrics that had 

 by Meta founder and CEO Mark Zuckerberg to enhance engagement on the platform and thus its profitability.

While Meta has publicly claimed to invest heavily in content moderation including by working with partners with local knowledge and expertise to catch violating content, Ethiopian members of its trusted partner program claimed that the corporation acted complacently when problematic content was reported. One partner reported that Facebook staffers responded that they are not “

 Meta would get rid of fact checkers in the US so as to “dramatically reduce the amount of censorship on our platforms” and “work with President Trump to push back on governments around the world that are going after American companies and pushing to censor more.”

Meta is being sued in Kenya because Facebook’s content moderation work for Ethiopia was being carried out in Kenya. This work was done through Sama, a business process outsourcing firm providing data labeling, computer vision training, and content moderation services. Sama refers to itself as an “ethical” AI outsourcing company which provides “

” to tech laborers. Sama claims to have “helped lift more than 65,000 people out of poverty,” many of whom got their first formal employment contracts from the corporation. However, a 2022 

 revealed “a workplace culture characterized by mental trauma, intimidation, and alleged suppression of the right to unionize.” Sama content moderators, who Facebook relied on to catch hateful content in Ethiopia, complained of pressure from managers to work fast, with requirements to deal with a ‘ticket’ (flagged item) in as little as 50 seconds or less.

One of the Facebook content moderators interviewed by Time complained that they lived “hand-to-mouth,” that they couldn’t “save a cent” and that the need to take care of their child was the only reason they didn’t resign. Youth unemployment rates in Kenya 

, forcing thousands to work under poor working conditions for lack of options. The fact that Sama brings desperately needed jobs into the country appears to be why the Kenyan government is quick to take the side of both Sama and Meta over its citizens, despite the complaints of Sama’s workers about poor pay and working conditions.

In March 2024, Kenyan President William Ruto met Nick Clegg, then Meta’s president of global affairs at the State House, and 

 for “offering to consider Kenya for some of the BPOs” (business process outsourcing firms) and offered assurance that he was “reworking the legislation in Kenya” to make it much easier for for firms in that sector to be able to employ young Kenyans. A few months later, in November 2024, Senate Majority Leader Aaron Cheruiyot proposed the 

 in the Senate, which would remove litigation risk for companies such as Meta when they worked with BPOs in the country. Defending the bill 

, Cheruiyot claimed that the proposed change was a response to “the players in the BPO space. A growing sector that currently employs thousands with the potential to explode and employ millions.” While the bill does not help firms such as Sama by reducing the risk of litigation by workers, it does help them avoid the termination of contracts by corporations such as Facebook once they are embroiled in legal battles. The bill was 

 in August this year, and now awaits presidential assent.

that remittances from exported labor function as subsidies for a state that has abdicated its developmental responsibilities and uses remittances as a subsidy to its role in citizen welfare and social reproduction. One might argue that a similar logic applies to the Kenyan government’s acquiescence to the wages and working conditions by BPO firms in Kenya. These wages, regardless of the oppressive conditions they are produced under (similar to 

Gulf remittances, produced under oppressive conditions

), subsidize the fiscal recklessness of a ruling class that uses state machinery for self-enrichment and has abandoned it as an object of collective transformation. As Africa is a Country editor William Shoki writes, such states are 

. They imagine foreign investors as their key stakeholders and for their citizens, only acting as an ‘interpreter of necessity’, explaining why they cannot act in line with ways that dignify the lives of their citizens, because otherwise, the foreign capital that subsidizes their abdicated responsibilities would leave. 

It remains to be seen how the case by Meareg, Tekle and the Katiba Institute will unfold. One can only hope that a rare instance in which the interests of Africans are upheld will manifest, offering an important lesson to Meta that African states are indeed willing to protect the rights of their citizens and operations on the continent will not be subjugated to corporations’ interests.

Jared Sine is the chief strategy & legal officer at GoDaddy.

The internet used to be something we navigated ourselves. Increasingly, AI systems are navigating it for us.

Millions of people already rely on chatbots like ChatGPT and Claude to answer questions, recommend products, summarize news, and help make decisions. Soon, AI agents like Meta’s Muse may compare insurance plans for users, schedule their doctor's appointments, 

For many, the convenience will be undeniable. But so will the power these systems have to steer attention, shape purchasing decisions, and determine which businesses thrive online.

The question is whether this new AI-powered internet will remain open to anyone with a good idea — or whether access to customers and visibility online will depend on the preferences of a few dominant platforms.

We are already beginning to see how this dynamic could evolve. Over the last few years major AI firms have 

with some of the world's leading news organizations. In some cases, these agreements help ensure those outlets appear prominently in AI-generated responses.

These partnerships have the potential to benefit everyone involved. Publishers gain compensation and visibility in an online ecosystem increasingly mediated by AI assistants. AI platforms gain broad access to trusted content and recognizable brands.

In other words, AI companies are beginning to occupy the role search engines once did — deciding which information users encounter first. But AI agents may ultimately wield even greater influence, since they won't just recommend options. They'll act on them.

By 2030, purchases "initiated, influenced, or completed" by AI agents could account for hundreds of billions in US e-commerce activity, 

If those agents favor companies that pay for preferred placement, sign licensing deals, or integrate directly with their platforms, a small number of AI firms could become the primary gatekeepers between consumers and the broader internet.

Businesses may find themselves competing not simply for customers but for placement inside a handful of closed AI ecosystems — much as companies once fought for shelf space in grocery stores or visibility in smartphone app stores.

The internet largely avoided this kind of gatekeeping in its early years because of how it was built. Shared technical standards ensured that no single company controlled who could participate online. Any business could create a website, register a domain name, and reach users across the web.

One of the most important of those standards was the Domain Name System, or DNS, which created a universal naming system for websites. Browsers did not decide which legitimate websites users could reach — nor did they charge businesses for the right to be discovered online.

The result was an internet where small startups could compete alongside giant corporations.

 would work much like DNS does for websites today. Rather than forcing businesses to negotiate access through a handful of dominant platforms, it would allow AI systems to discover and interact with businesses and services across an open network, regardless of which platform those agents run on.

Instead of operating inside a few closed platforms, AI agents could interact across a more open and competitive network — much as browsers and websites do today.

Several organizations are already developing open frameworks for agent identity, interoperability, and authentication. For instance, Google's 

, now under the Linux Foundation, allows AI agents built by different companies to communicate with one another. Anthropic's 

, also under the Linux Foundation's umbrella, provides a common way for AI systems to connect with outside tools and data sources.

Whatever form these efforts ultimately take, the principle behind them is critical — the next generation of the internet should be shaped by open standards, not private permission slips.

The openness that defines today's internet did not happen by accident. Its architects built systems that distributed power instead of concentrating it.

As AI reshapes how people interact with the digital world, we now face a similar choice. We can preserve the internet as an open marketplace, or allow a handful of platforms to become its new toll booths.

Executive Vice-President for Tech Sovereignty, Security and Democracy of the European Commission Henna Virkkunen, and European Commissioner for Democracy, Justice, the Rule of Law and Consumer Protection Michael McGrath speak at a press conference on Wednesday. 

This morning, the European Commission published its 

 for the Digital Fairness Act (DFA), a major initiative to update the European Union’s consumer protection framework for an increasingly digital economy. The proposal seeks to address unfair commercial practices that exploit consumers through manipulative online interfaces and addictive design features and other forms of digital commercial conduct.

This proposal translates many valuable ideas into legal language, but questions remain about its scope, coherence and capacity to deliver effective protection.

In this commentary, we look at five central aspects of the proposal, including its title and scope, the move from a directive to a regulation supported by delegated acts, the prohibition of dark patterns, the regulation of addictive design features, and the new rules regarding enforcement. We identify significant advances, but also areas where the legislative text requires clarification or substantive improvement.

The coming weeks will be an opportunity to carefully consider additional aspects, such as respect for the principle of technological neutrality and whether the proposal can deliver a high level of consumer protection in relation, for example, to AI agents and price personalization.

The title is a misnomer. For years, the EU has promised to confront the business models and design practices that exploit consumers in digital markets. The title 

 reflects that agenda, but understates the scope of the proposal: the DFA is not about digital services only. If adopted, the proposal would repeal the 

Unfair Commercial Practices Directive (UCPD)

 and replace it with a regulation governing B2C commercial practices, both online and offline. A more accurate title would be 

Leveraging the expressive function of the law does not justify starting a tradition of misnamed legislative instruments. The Digital Services Act (DSA), for example, does not cover 

 digital services. Similarly, the DFA obscures unfair practices that happen in real life. There is little justification for presenting a reform of consumer law with such broad implications as exclusively concerned with digital markets. The politically important phrase ‘digital fairness’ gives the title to Chapter III, the proposal’s most innovative part. Additional emphasis, if deemed useful, could be given in the recitals.

From a directive to a regulation supported by delegated acts

A regulation is directly applicable. It may reduce interpretative fragmentation and facilitate stronger cross-border enforcement, particularly if accompanied by the proposed reform of the 

Consumer Protection Cooperation Regulation.

 The draft consolidates interpretations of the UCPD that have emerged over the years, for example, clarifying that continuing to use a service constitutes a ‘transactional decision’ (Article 3(3)). Another example is the recognition that vulnerability is not a matter of categories (age, disability) but can result from personal circumstances (Article 5(3)).

These clarifications matter because they translate developments in the interpretation and application of consumer law into a more accessible and consistent legislative framework.

The choice of giving the Commission updating powers through delegated acts is welcome. In 

, we already advocated such a mechanism. However, the scope of this power forgets what happens in real life. The Commission should have the power to update the whole blacklist, not just the section on digital practices. Moving this delegation from Articles 10 and 17 to Article 5 would ensure that the UCPR is future-proofed in a technology-neutral manner.

Article 10 of the proposal prohibits dark patterns, giving legislative recognition to a concern raised for years. This is welcome. The widespread use of manipulative design online has made it increasingly important to clarify how consumer law applies to these practices and to make those rules easier to enforce.

The case for a standalone prohibition is less clear.

Article 5 qualifies violations of Article 10 as unfair commercial practices. Article 10(1), just like Article 25(1) DSA, provides less guidance than Articles 5-9 UCPD/DFA. In our 

, we identified 42 dark pattern types and showed that they all fall within existing UCPD categories (misleading actions, omissions and aggressive practices or the general prohibition). The problem was never that consumer law did not prohibit dark patterns or lacked the fashionable expression “dark patterns”. It was whether its established concepts could lead to effective enforcement against the widespread use of unfair commercial practices online. That was and still is the goal.

The relationship with the DSA remains troubling. Article 25(2) DSA already excludes practices covered by the UCPD or the GDPR. 

, we argued that this provision reverses the ordinary logic of general and special legislation: rather than refining consumer-law protection, the sector-specific DSA rule risks ceasing to apply whenever consumer law can be invoked. The draft does not repeal Article 25(2) DSA as many had hoped. Instead, in a masterful act of make-believe, Recital 52 declares that the DSA prohibition does not apply to dark patterns covered by the DFA, which ‘relate exclusively to business-to-consumer commercial practices’. One small problem: the DSA prohibition also covers business-to-consumer commercial practices occurring within online interfaces. Thus, the circumstances in which Article 25(1) DSA remains applicable are still an elusive mystery.

Article 17 of the proposal breaks new ground by recognizing features that are intended, or can reasonably be foreseen, to encourage compulsive or excessive use. It lists uninterrupted consumption without breaks and autoplay, engagement-triggering notifications, rewards and penalties for returning, and simulated interpersonal relationships. This is an important clarification: such problematic practices could not be disciplined well only by the elusive notion of undue influence.

However, the proposed regulatory response falls short of the problem it identifies.

Rather than prohibiting addictive configurations for adults, Article 17 requires tools that let consumers switch them off. The Commission is not asking to remove the problematic features: the consumer must choose to disable them. This approach sits uneasily with Recital 67, which recognizes that addictive design exploits consumers’ cognitive and psychological vulnerabilities.

The Regulation acknowledges those vulnerabilities while relying on consumers to recognize and counteract their exploitation. It is easy to imagine that traders would make the opted-out version so cumbersome that consumers will choose to remain with the addictive design. Moreover, consumers have been using these addictive features for years and have grown accustomed to them to exercise self-control and opt out of them. The right to opt out is too weak a response in the face of many years of harmful exposure. It is not even complemented by a risk to inform about the risks associated with these problematic features.

 makes the difference in regulatory approaches more visible. For certain services used by minors, it envisages stronger safety-by-design restrictions (KIDS ACT, Chapter III). This divergence in treatment will require strict scrutiny through the legislative process.

Strengthened public powers, but what about effective remedies?

Perhaps the most important innovation is found in Article 30: the Commission will have direct enforcement powers for infringements in the EU. Most likely, DG JUST will join the club of DGs with enforcement powers. The Commission's extensive experience in interpreting the UCPD and the 

 (UCTD), reflected in its detailed guidance notices, provides a strong foundation for exercising these new enforcement powers.

However, penalties and consumer compensation remain a weakness.

Stronger enforcement powers by the Commission will achieve little without sufficiently deterrent penalties and effective consumer remedies. Leaving Member States to define the amount of fines on their own seems undesirable, as this New Mexico-Italy comparison shows.

, a state with 2.1 million inhabitants secured a $ 25 billion judgment against the company. The court calculated a potential penalty of approximately $219.5 billion by applying the statutory maximum of $5,000 to each of the violations found by the jury. It subsequently reduced the award to $25 billion.

It did so with the functional equivalent of the UCPD. How? One such statement led to a 10 million EUR 

 under the UCPD. The contrast shows how important dissuasive sanctions are. Old legal principles can be powerful when backed by evidence, institutional resolve and effective remedies.

At the same time, private enforcement needs more. It is relatively easy to find a commercial practice unfair, much less so to prove the damage consumers suffered. Thus, Article 34 should be reworded to grant explicit evidentiary support also concerning the harm caused by the relevant commercial practice. Moreover, as 

, indemnities based on a percentage of the price, or at least a fixed amount, which are common in EU passenger regulations, or other benchmarks (e.g., turnover) are crucial to make consumer rights effective in most cases of mass harm.

The DFA is an important opportunity to strengthen the EU’s response to unfair commercial practices. It remains to be seen whether the Parliament will be able to amend the text as needed.

 accompanying the DFA proposal cites several academic publications by the authors of this article.

View at the Google DeepMind logo at the office building in London. (AP Photo/Alastair Grant)

The Google Whistleblower Taking on Big Tech Over Its Role in Gaza

Exterior view of the US headquarters of TikTok in Culver City, California, in 2022. Shutterstock

Exterior view of the US headquarters of TikTok in Culver City, California, in 2022. 

Social Media Addiction Cases Are Settling Before Trial. What Does It Mean?

A crowd of supporters surround President Lula holding a microphone at a podium draped with the Brazilian flag.

President Luiz Inacio Lula da Silva delivers a victory speech after winning the presidential runoff in Brazil, Oct. 25, 2026. (AP Photo/Andre Penner)

Lula’s Win Signals Fight to Regulate Social Media Will Escalate in Brazil

Tech Policy Press - Technology and Democracy

Tech Policy Press is tracking laws and regulations, along with government investigations and litigation, that will shape the rules and accountability for tech companies.

Latest

Podcast

Recent

Our content delivered to your inbox

Thank you!

Around the globe