A draft of the EU’s Digital Fairness Act would target AI systems that simulate personal relationships to encourage users to keep interacting with them, Bloomberg News reported on Monday. Companies found to use unfair digital trade practices would risk fines of as much as 6% of annual global turnover, Reuters reported from a Commission draft it also saw.
That 6% is 1.5 times the level EU consumer law set in 2019, by our arithmetic. The base changes too: the 2019 directive sets the maximum fine at no less than 4% of turnover in the EU countries concerned, while the draft counts global turnover, according to Reuters.
Business Pill · Consent
An evergreen Business Engineer explainer on consent: what it means for a company to have a user’s permission. It explains the general idea only and does not discuss the Digital Fairness Act or any company in this piece.
The key insight: The reported draft moves engagement design into consumer law. Bloomberg says it targets AI systems that simulate personal relationships to keep users interacting, and Reuters says fines could reach 6% of annual global turnover, which is 1.5 times the 2019 consumer-law level, by our arithmetic.
Bloomberg says the draft would require companies to give adult users the ability to switch off features that encourage compulsive or excessive use of their services. In Bloomberg’s words, “the measures would also target AI systems that simulate personal relationships to encourage users to continue interacting with them”.
The part of Bloomberg’s report we could read does not say whether those AI measures are a ban, a design limit or a disclosure duty. Investing.com, relaying the Bloomberg report, describes the target as conversational AI designed to drive compulsive engagement.
Maximum fines as a share of annual turnover under EU laws in force (EUR-Lex) and under the draft Digital Fairness Act as reported by Reuters on 9 November. The draft has not been published; the 1.5 times comparison is our arithmetic.
Switch-Offs, Dark Patterns and Ads Aimed at Children
Reuters lists the rest of the package from the draft. It would ban business-to-consumer “dark patterns”, which Reuters defines as deceptive or manipulative interface designs that steer people into buying more, as unfair commercial practices. It would also ban personalised advertising aimed at children or vulnerable consumers.
Traders would have to let users switch off addictive features that encourage compulsive or excessive use, the draft says, according to Reuters. Sellers using dynamic pricing would have to disclose how many items are available at the advertised price.
From 4% of EU Turnover to 6% of Global Turnover
The 2019 directive tells EU countries to set the maximum fine for widespread consumer-law breaches at “at least 4 % of the trader’s annual turnover in the Member State or Member States concerned”. The Reuters figure for the draft is 6% of annual global turnover.
That puts the draft level with the Digital Services Act, which caps fines at 6% of annual worldwide turnover, and above the GDPR’s 4% of worldwide turnover for the breaches listed in its Article 83(5). The AI Act goes higher, at 7% of worldwide turnover for its prohibited practices.
Those AI Act prohibitions already cover AI that uses “purposefully manipulative or deceptive techniques” or exploits vulnerabilities linked to age, but only where the behaviour causes or is reasonably likely to cause significant harm, according to Article 5 of the AI Act. The part of the draft we could read, through Bloomberg, mentions no such harm test; the full text has not been published.
Wednesday’s Presentation
Justice Commissioner Michael McGrath will present the proposal on Wednesday, Reuters reports, and it will then need to be negotiated with EU countries and the European Parliament before it can become law. The Commission listed the Digital Fairness Act in its 2026 work programme for the fourth quarter, according to the European Parliament’s legislative tracker.
The tracker says a Commission consultation from July to October 2025 showed differing views: consumer groups called for tighter rules, while business representatives favoured better enforcement of existing law. Investing.com, citing Bloomberg, says EU policymakers are separately preparing a broader “Kids Act”.
As we read the two reports, the draft puts relationship-simulating AI in the same frame as addictive features: both are treated as design choices that keep users engaged, and both are handled as consumer-law questions.
Bloomberg News, reporting on a draft of the EU Digital Fairness Act, 9 November 2026
“The measures would also target AI systems that simulate personal relationships to encourage users to continue interacting with them”
Three Implications
FOR AI COMPANION AND ASSISTANT APPS Bloomberg says the draft targets AI systems that simulate personal relationships to encourage continued interaction; the part of the report we could read does not say what form the rule takes.
FOR PLATFORMS WITH EU USERS Reuters says traders would have to let users switch off addictive features, and that personalised ads aimed at children or vulnerable consumers would be banned.
FOR THE TIMETABLE Reuters says McGrath presents the proposal on Wednesday, after which EU countries and the European Parliament must agree on it before it can become law.
What Is Not Established
The draft has not been published, and the Commission had made no public statement on it when this piece was written. This piece rests on the two opening paragraphs of Bloomberg’s report, which is otherwise paywalled, and on Reuters’ full report. It is not established whether the 6% ceiling applies to the AI-bot measures, which AI products would fall in scope, or whether the text changes before Wednesday.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
As reported by Bloomberg and Reuters, the draft treats engagement design itself, from addictive features to AI that simulates relationships, as a consumer-law issue. Its reported fine ceiling of 6% of global turnover is 1.5 times the 2019 consumer-law level, by our arithmetic, and is measured on a wider base.
A note on sourcing. The AI-bot and switch-off points come from the readable opening of Bloomberg’s 9 November report; the 6% fine level, the dark-patterns and advertising points and Wednesday’s presentation come from Reuters’ report of the same day. Both outlets saw a Commission draft that has not been published. Fine levels in force are quoted from the laws on EUR-Lex, and the timeline from the European Parliament’s legislative tracker. The 1.5 times comparison is our arithmetic.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
