Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
On July 2, 2024, Brazil’s National Data Protection Authority (ANPD) ordered Meta to stop processing personal data from its products to train generative-AI systems. The preventive order also suspended the related privacy-policy language and imposed a daily fine of R$50,000 for noncompliance. It was not a permanent ban on Meta AI or Facebook, Instagram, or Messenger: ANPD later approved a restricted restart on August 30, 2024.
The short timeline
- June 26, 2024: Meta’s updated privacy policy took effect, authorizing the use of publicly available information and content shared on its platforms to train and improve generative-AI systems.
- July 2: ANPD issued a preventive suspension covering the relevant policy language and data processing in Brazil.
- July 10: The authority maintained the measure after Meta requested reconsideration.
- July 17–18: Meta paused generative-AI features that were active in Brazil while it engaged with the regulator.
- August 30: ANPD approved a compliance plan and allowed processing to resume under restrictions.
The initial order and later decisions are documented in ANPD’s July 2 notice, its July 10 reconsideration decision, and its August 30 compliance decision.
What Brazil actually halted
ANPD’s action concerned three related but distinct matters:
- Meta’s new privacy-policy language about using personal data for AI training;
- the underlying processing of personal data for training generative-AI systems; and
- certain user-facing generative-AI features that Meta later paused in Brazil.
The order applied to Meta products including Facebook, Instagram, and Messenger. It did not shut down those services, ban Meta AI worldwide, or prohibit all AI training. Nor did it establish that every Brazilian user’s data had been used. It addressed the policy’s authorization and the potential scope of the processing.
#1 Best Overall
ANPD said the relevant material included publicly available information and content shared through Meta’s products. That could include posts, photographs, videos, comments, and other material, including older content. The authority also said the processing could involve information about people who were not Meta users—for example, a non-user appearing in another person’s public photo or post.
The cited material does not establish that private Messenger conversations were included, so it would be inaccurate to describe the order as a finding about private messages.
Why ANPD intervened
Brazil’s Lei Geral de Proteção de Dados Pessoais (LGPD) does not automatically make public information free for any later commercial use. ANPD’s analysis treated public posts as personal data that remains subject to legal principles including good faith, purpose limitation, necessity, and transparency. Its later technical note described AI training as a secondary use of data originally collected for other purposes.
Recommended Free Tools
The regulator identified preliminary indications of several problems:
- Legal basis: ANPD questioned whether Meta’s asserted legal basis was appropriate for this processing.
- Transparency: The policy change may not have been presented in a sufficiently clear, prominent, and accessible way.
- Objections: Users could face excessive practical obstacles when trying to object, such as finding the relevant information, understanding the choice, completing the process, or receiving confirmation.
- Purpose and expectations: People may have posted for communication with friends, communities, or businesses—not with the expectation that their material would later be aggregated into generative-AI training datasets.
- Minors: Photos, videos, and posts involving children and teenagers raised additional concerns under the LGPD’s best-interest requirements.
- Sensitive information: Public images or text can reveal health, religion, political views, race, sexual orientation, location, or other sensitive details.
ANPD cited the potential scale as well: it said Facebook had approximately 102 million active users in Brazil at the time of the 2024 decision. That is a dated figure, not a current user count.
A preventive measure—not a final ruling
The July order was a preventive measure based on preliminary findings and ANPD’s assessment that continued processing could cause serious or difficult-to-repair harm. It was not a final adjudication that Meta had definitively violated the LGPD.
That distinction matters for AI systems. Once personal data has been incorporated into training pipelines, reproducing the original dataset, identifying every downstream artifact, or excluding the effect of particular data can be technically and legally difficult. ANPD therefore acted before completing the underlying supervisory proceeding.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Meta’s response and the feature pause
Meta’s position, reported by Reuters, was that its approach complied with Brazilian law and that it would continue working with ANPD to address the authority’s questions (initial report). After ANPD maintained the measure on July 10, Meta said it would pause generative-AI tools already active in Brazil while it addressed the regulator’s concerns (Reuters; TechCrunch).
Rank #3
The R$50,000 figure was a daily penalty for failing to comply with the preventive order, not a declaration that Meta had already been fined that amount as a final privacy-law sanction.
What changed on August 30
ANPD later suspended its prohibition after approving Meta’s compliance plan. The restart was conditional, not unrestricted. The plan required clearer information for people affected by the processing, a simpler and more practical objection mechanism, and a notification period. It also committed Meta not to use data from accounts belonging to people under 18 while the supervisory process continued.
In its November 11, 2024 technical assessment, ANPD said the proposed safeguards were sufficient at that point to improve transparency and facilitate objections, while continuing to identify purpose-limitation and transparency questions about the secondary use of public Facebook and Instagram data.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsWhat this means for users
Can people still use Facebook and Instagram?
Yes. The order did not suspend Meta’s social networks. It targeted a specific personal-data processing activity and related policy language.
Was every post automatically used for training?
Not established. The policy authorized a potentially broad category of processing; the available decisions do not prove that every Brazilian user’s content entered a training dataset.
Did the action cover non-users?
Yes, in the sense that ANPD’s description covered personal data processed through Meta products even when the person described or pictured was not a Meta account holder.
What about children and teenagers?
Their data was a central concern. Under the approved compliance plan, Meta committed not to use data from accounts belonging to people under 18 while the supervisory process continued.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Is the case finished?
The sources available for this article verify the preventive suspension, its confirmation, and the restricted restart. They do not verify a later final merits decision closing the underlying supervisory process. The lifting of the prohibition should therefore not be described as either a permanent exoneration or a permanent ban.
What can a concerned user do?
Users should read Meta’s current privacy notices, look for the company’s applicable objection process, and keep confirmation of any request. Because interface labels can change, this article does not prescribe a current menu path. Complaints about data processing can also be directed to Brazil’s ANPD.
Why the case matters beyond Brazil
The dispute illustrates a broader question in AI governance: does making information public authorize a company to repurpose it indefinitely for model training? ANPD’s approach says the answer cannot be assumed. Regulators may examine whether the new use is compatible with the original context, necessary, clearly explained, and subject to meaningful control.
It also shows a middle path between unrestricted processing and a permanent prohibition. A regulator can pause a high-risk use, require operational safeguards, and later permit a restricted restart. The trade-off is real: delays may reduce access to AI features or the availability of Brazilian-language and culturally specific training material, but allowing irreversible processing without adequate controls can make later remedies ineffective.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Status as of August 16, 2026: The July 2024 suspension was lifted on August 30, 2024 under a restricted compliance plan. The sources reviewed here do not verify a later final merits decision resolving the underlying supervisory process.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
