European regulators have initiated an antitrust inquiry into Meta after WhatsApp banned third-party AI chatbots from its Business API while making the company’s own Meta AI available to users. The European Commission has said it is worried the policy will block rival AI providers from an important distribution channel in the European Economic Area, leaving a fast-growing market to swing Meta’s way.
What Triggered the EU Probe Into WhatsApp’s AI Policy
WhatsApp shifted its Business API rules in October, sharing that the API was not designed to be a surface area for general-purpose AI assistants. The new terms, which are set to take effect in January, would block tools like ChatGPT-style bots from running within WhatsApp’s systems even as Meta introduces its own Meta AI assistant on the app.
The policy does not prevent firms from using AI to power customer support or transactional messaging. An automated help bot for a retailer would still be permitted. But non-company-specific AI chatbots would be cut off from a channel that is for many users their default communications platform.
The Commission’s worry is simple and direct: banning rivals from reaching hundreds of millions of European users who hold WhatsApp in their grip every day, while Meta — through its own virtual assistant — still gets to stay inside the product. Officials are looking at whether that arrangement constitutes exclusionary conduct and self-preferencing by a dominant messaging platform.
Why Competition Authorities Care About WhatsApp Access Rules
WhatsApp’s popularity in Europe means distribution decisions on the app have consequences. Consumer polls consistently find that WhatsApp’s penetration exceeds 80% in some of the largest EU markets, such as Spain and Italy, and more than 70% in others, like Germany and the U.K. WhatsApp, which has more than 2 billion users worldwide, is the main portal to digital communication for huge numbers of people.
In markets where discovery and day-to-day usage rely on messaging apps, losing access to APIs could substantially increase customer acquisition costs for independent AI chatbots while reducing user options. The Commission will probe if Meta is using the leverage in messaging to also preference its own AI assistant — not an issue which might firmly engage EU abuse-of-dominance rules (given how it’s couched under Article 102 TFEU).
There is also the Digital Markets Act, which establishes that large platforms are gatekeepers and bans self-preferencing and discriminatory access. While the current action is antitrust-focused, the DMA objectives — ensuring fair terms for business users and guaranteeing contestability — form an element of our policy on bottleneck control in strategic markets such as messaging.
Meta’s Defense and Architectural Shortcomings
Meta says the Business API wasn’t designed to support or distribute free-floating AI assistants, and accommodating them would strain systems built for customer support and notifications. Consumers can still discover competing chatbots in app stores, search engines, email, partnerships and device integrations, the company says: no foreclosure.
The competition enforcers will be examining whether those alternate pathways are comparable to distribution within WhatsApp, where engagement is habitual and notifications are native. In antitrust or competition terms, the challenge will be to Meta’s argument of “technical necessity:” Is the restriction also objectively justified and proportionate, rather than a crude but convenient rule that shelters its own assistant?
Scale matters. Meta says millions of businesses depend on WhatsApp to chat with customers, and in many countries it is a default channel for small merchants. And denying general-purpose AI bots access to that surface could shape which tools consumers actually try — especially given early user behavior for generative AI remains fluid, and discovery friction has an outsize effect.
What Could Happen Next in the EU’s Investigation of Meta
The Commission has a broad toolkit. If it uncovers immediate dangers, it can pursue provisional measures to avoid irreparable damage amid the investigation. In the end, it can fine up to 10% of Meta’s worldwide turnover and order behavioural remedies (like a nondiscriminatory treatment with API access, transparent criteria of access or technical capabilities for vetted AI assistants).
There are other routes with commitment offers: Meta might promise voluntary remedies to deal with any worrying aspects, without an analysis of abuse. Not just the competition on the merits, but the wider signal to gatekeepers that issue rules for AI deployment across their ecosystems.
The stakes extend beyond WhatsApp. Messaging platforms have turned into operating systems in their own right — from Telegram’s bots to WeChat’s mini programs — and rules can make or kill entire new categories. In Europe’s opinion, as AI assistants seem to be getting in position for a new consumer-facing interface layer, it makes a mistake by not giving itself access terms on dominant platforms — such that are formulated strictly according to the platform’s own commercial axioms.
For AI startups, the result will dictate go-to-market strategy in Europe: whether WhatsApp stays a no-go area, opens up with tight technical guardrails in place, or becomes a neutral conduit for third-party assistants. For users, the pragmatic question is more straightforward — will there be variety of choice in chatbots where they’re already spending time? The goal of the Commission’s investigation is to ensure that this question isn’t answered simply based on who owns the platform.
