In a striking challenge to the authority of a global technology titan, Brazil’s primary competition authority has decisively intervened to prevent Meta Platforms from implementing a policy that would effectively banish third-party artificial intelligence chatbots from its dominant WhatsApp messaging service. Just days before the revised terms of service were scheduled to take effect worldwide on January 15, 2026, the Administrative Council for Economic Defense (CADE) imposed interim measures, temporarily suspending the policy and raising profound questions about a digital gatekeeper’s right to leverage its market power in one sector to control the future of another. This bold action, backed by the threat of substantial daily fines, represents a critical flashpoint in the burgeoning battle between regulators and Big Tech over the competitive landscape of the nascent AI industry. The move, which closely mirrors a similar preemptive strike by Italian regulators, suggests a growing international resolve to address potentially anticompetitive behavior before it can irrevocably shape the market, setting the stage for a prolonged legal and regulatory confrontation with global implications.
The Policy and the Pushback
The heart of the dispute lies in a modification to the WhatsApp Business Solution Terms, first unveiled by Meta in October 2025, which explicitly prohibits AI service providers from accessing the platform if their core function is to offer AI capabilities. This seemingly subtle change in contractual language was poised to have a seismic impact, severing the connection between a new generation of general-purpose AI assistants and WhatsApp’s nearly three billion users. The ban would have immediately impacted major global players like OpenAI’s ChatGPT, Microsoft Copilot, and Perplexity, as well as burgeoning regional providers that had successfully built a user base on the platform, such as Spain’s Luzia and Colombia’s Zapia. These services, which had been operating and growing within the WhatsApp ecosystem, suddenly faced an existential threat that could erase their primary channel for user engagement and data collection, effectively ceding the entire market to Meta’s own integrated solutions.
In a swift and decisive response, CADE issued a detailed technical note outlining its order against Meta Platforms, its subsidiary WhatsApp LLC, and its Brazilian arm. The primary component of the order was the immediate suspension of the controversial contractual provisions, freezing the policy change until a full administrative investigation could be concluded. This measure effectively preserved the competitive status quo, allowing third-party AI providers to continue their operations without interruption. More importantly, the order included a crucial preventative clause, explicitly forbidding Meta from deploying any alternative contractual, technical, or operational measures that would achieve the same exclusionary outcome. This forward-thinking provision was designed to prevent the company from circumventing the ruling through subtle workarounds, ensuring the spirit of the decision was upheld while the deeper investigation proceeds. To ensure compliance, CADE attached a substantial daily penalty of R$250,000 (approximately $50,000), a figure calibrated to make defiance an economically untenable option for the technology giant.
CADE’s Case Against Meta
The legal foundation for CADE’s intervention is a robust preliminary analysis that uncovered compelling evidence of anticompetitive conduct by Meta. Central to its argument is the undisputed fact of WhatsApp’s overwhelming market dominance in Brazil. The authority’s investigation presented stark statistics to support this conclusion: WhatsApp is installed on 99% of smartphones in the country, serves a staggering 150 million users, and is accessed daily or almost daily by 97% of them. This near-total market saturation, reinforced by powerful network effects that make it nearly impossible for a competitor to gain a foothold, grants Meta a degree of market power that invites heightened regulatory scrutiny. The sheer scale of this dominance means that any action taken by Meta to restrict access to its platform has profound and immediate consequences for competition in adjacent markets that rely on its infrastructure for distribution and user engagement.
Building on this premise of dominance, CADE’s legal experts identified Meta’s strategy as a classic example of “offensive leveraging.” This antitrust theory describes a situation where a firm weaponizes its power in a mature market—in this case, instant messaging—to foreclose competition and secure an unearned advantage in an adjacent, emerging market like AI services. The authority meticulously detailed what it saw as a calculated, multi-stage plan by Meta. First, the company permitted and even welcomed third-party AI providers onto its platform, which fostered user adoption and normalized the use of AI assistants within the WhatsApp environment. Second, after the market was primed, Meta introduced its own competing service, Meta AI, granting it preferential integration and access. Finally, in the third and most aggressive step, it moved to completely exclude all external competitors, effectively clearing the field for its own product just as the market was beginning to mature. This sequence of actions, CADE argued, was not indicative of competing on the merits but of a deliberate strategy to monopolize a new technological frontier.
The Brazilian authority further concluded that a total exclusion of third-party AI tools was a disproportionate and unjustifiable measure. Invoking the “special responsibility” doctrine, a legal concept adapted from European Union competition law, CADE asserted that a dominant firm has a heightened obligation to pursue its business objectives through the least restrictive means possible. The fact that Meta’s own AI service continues to operate seamlessly within WhatsApp served as powerful evidence that the platform could, in fact, support such services without issue. This undermined Meta’s technical justifications and suggested that less anticompetitive alternatives to a complete ban—such as setting clear technical standards, implementing fair commercial terms, or developing a certification process—were not only plausible but obligatory for a firm with such market power. The urgency of the intervention was also a critical factor. CADE determined that allowing the ban to take effect on January 15 would inflict serious and likely irreversible competitive harm. Once third-party AI providers were severed from their primary user base, the market dynamics would shift decisively and perhaps permanently in Meta’s favor, a situation that subsequent enforcement actions might be unable to remedy.
Meta’s Defense and the Rebuttal
In its formal response to CADE on December 15, 2025, Meta presented a series of justifications for its policy change, attempting to frame it as a necessary business decision driven by technical and commercial imperatives rather than a calculated anticompetitive strategy. The company’s primary technical argument centered on the claim that general-purpose AI assistants were causing significant and unforeseen infrastructure problems. Meta contended that these services generated message volumes that far exceeded the typical customer service interactions the WhatsApp Business API was designed to handle, leading to system failures on at least three separate occasions. This excessive use was characterized as a form of “free-riding,” where AI developers were allegedly exploiting Meta’s infrastructure at a massive scale without providing appropriate compensation, thereby jeopardizing the stability and performance of the platform for all users.
Beyond the technical concerns, Meta argued that the policy was a legitimate business decision intended to protect its substantial investments in developing its own paid commercial messaging services. The company sought to draw a key distinction, emphasizing that its restrictions were not meant to prevent businesses from using AI to enhance their own customer communications. Instead, the ban was specifically targeted at entities whose core, primary service was offering a standalone AI assistant on the platform. From Meta’s perspective, this was a necessary step to define the intended use of its business tools and to create a sustainable commercial model that would fund future platform development. By positioning the ban as a targeted measure to protect its business model, Meta attempted to portray its actions as reasonable and within the bounds of standard commercial practice, rather than an abuse of its dominant market position.
However, CADE found these justifications to be unpersuasive at the preliminary stage of the investigation. The authority determined that Meta’s arguments related primarily to the merits of the case, which would require a full and thorough procedural investigation to validate, rather than addressing the fundamental and immediate anticompetitive nature of the conduct. Regarding the urgency and potential for harm, CADE systematically dismantled Meta’s claims. It found no credible evidence that suspending the policy would damage WhatsApp’s operations, as the suspension merely preserved the existing operational conditions that had been in place for months. CADE astutely pointed out that Meta itself had allowed these AI assistants to operate for a significant period after announcing the ban in October and had even provided a three-month grace period for user migration. The authority reasoned that if the infrastructure problems were truly as critical as claimed, Meta would not have permitted such an extended and leisurely transition period, suggesting the technical rationale may have been overstated.
The Growing Global Scrutiny
The Brazilian action against Meta is not an isolated event but rather a significant development within a broader, coordinated international wave of scrutiny targeting the company’s AI strategy. The move by CADE was preceded by a nearly identical step from Italy’s competition authority, Autorità Garante della Concorrenza e del Mercato (AGCM), which on December 24, 2025, imposed its own precautionary measures to suspend the WhatsApp Business Solution Terms modifications within Italian territory. The AGCM’s reasoning closely mirrored that of its Brazilian counterpart, identifying Meta’s conduct as a credible instance of offensive leveraging designed to extend its dominance from messaging to the AI sector. The Italian decision was so relevant that CADE extensively cited it in its own analysis, underscoring a shared concern among European and South American regulators about the potential for digital gatekeepers to stifle innovation in nascent technology markets.
Both the Brazilian and Italian authorities underscored the critical timing of Meta’s policy change. They argued that by excluding rivals at this crucial developmental stage, Meta could secure a lasting and potentially unassailable advantage in the AI market. With its own Meta AI product as the sole chatbot service operating within the platform, it would be the only one capable of training on the vast and unique dataset of WhatsApp user interactions. This exclusive access would create powerful learning advantages and network effects that competitors, relegated to other distribution channels, could never replicate. Over time, this would lead to high switching costs for users who become dependent on a single AI assistant deeply integrated into their primary communication tool, solidifying Meta’s control over the market before meaningful competition could even emerge. This shared concern highlights a proactive regulatory posture aimed at preserving market contestability during the formative years of a new technology.
This heightened oversight from national competition bodies is complemented by broader regulatory pressure. The European Commission has intensified its scrutiny, opening a formal investigation on December 9, 2025, into how digital gatekeepers may be establishing unfair competitive advantages in AI through privileged access to infrastructure and exclusive data partnerships. This investigation runs parallel to existing challenges Meta faces in Europe under the Digital Markets Act regarding its advertising model. These actions in Brazil and Europe stand in stark contrast to a recent outcome in the United States, where a federal judge dismissed the Federal Trade Commission’s monopolization lawsuit against Meta. That judge found that in the U.S. social media market, platforms like TikTok and YouTube provide sufficient competition to Facebook and Instagram. This highlights a crucial jurisdictional difference in market definition: while Meta faces robust competition in social media in the U.S., its WhatsApp service enjoys overwhelming dominance in the instant messaging market in Brazil and Europe, leading to fundamentally different legal conclusions and regulatory actions.
The Fragmented Future
The immediate and most direct consequence of the Brazilian and Italian orders was the geographic fragmentation of Meta’s global policy. As of January 15, 2026, the ban on third-party AI providers was implemented worldwide, creating a new set of rules for the AI industry, but with two glaring exceptions. This has created significant operational complexity for Meta, which must now maintain a dual policy framework, enforcing the ban in most of the world while permitting open access in two key international markets. This jurisdictional split also presents a challenge for AI providers, whose access to a potential user base of nearly three billion people is now dictated by national borders and the varying appetites of antitrust regulators, creating uncertainty and complicating global growth strategies.
For the competing AI services caught in the crossfire, the interim measures have provided a vital, if temporary, reprieve. They can continue to serve their users in Brazil and Italy, maintaining the crucial user relationships and data flows that are essential for training and improving their machine learning models. This preservation of access, during what could be a lengthy and complex investigation, may prove to be competitively decisive. It prevents Meta from gaining an insurmountable head start and allows these smaller players to continue innovating and competing on the merits of their technology. The breathing room afforded by these orders could be the difference between survival and market extinction for a new generation of AI companies.
Ultimately, the full administrative inquiry in Brazil, along with the parallel proceedings in Italy and the broader investigation by the European Commission, served as a crucial test case for applying traditional antitrust principles to the rapidly evolving AI landscape. The detailed analysis of Meta’s business terms, its economic justifications, and the competitive dynamics of the interconnected messaging and AI markets established a powerful precedent. The outcomes of these cases have begun to shape how competition law will address the leveraging of market power by dominant platforms in other adjacent digital markets, from advertising and payment services to cloud infrastructure. The decisions made in these formative battles have had a lasting impact, influencing the structure of the digital economy for years to come by defining the rules of engagement in an era of unprecedented technological integration.
