Copy article

New US Facebook Forum app promises discussion anonymity in support groups. The privacy policy says otherwise.

ended 27. May 2026

Meta's new US Forum app promises "deeper discussions" in support groups and niche communities, while its App Store privacy label declares it collects health data, sensitive information, precise location, and browsing history, all linked to your identity.

Forum landed in the App Store on 22 May with no official announcement, positioning itself as a dedicated space for Facebook Groups, stripped of the main feed's “noise”.  Users can post under nicknames. 

The pitch is community, authenticity, real answers from real people but the privacy label tells a different story. Forum collects 14 categories of data linked to a user's identity, including health and fitness, financial data, contacts, and precise location, across six declared purposes including third-party advertising and developer marketing. 

Admins can see the real identity behind every pseudonym. An AI "Ask" feature aggregates answers across groups; an AI admin assistant moderates content. 

Privacy experts wonder what will happen when people share their most personal concerns inside a product whose entire business model runs on making that data profitable. An iOS-only app means every user is on a device that provides precise location, health data from HealthKit integrations, and the full mobile sensor suite. A web version would let people access groups through a browser with ad blockers, container tabs, VPNs, the kind of friction that limits data collection.

  • Forum collects sensitive data, and precise location for an app marketed at support groups. Should Apple allow this app in its store?
  • A single data breach could map every pseudonym to a real Facebook identity and location. What might that mean for vulnerable users?
  • Meta sets the moderation rules that Forum's AI admin assistant enforces,what happens to community self-determination when one company controls the AI-based governance tool?
  • Tech companies make their websites clunky to use via a browser to force users onto the “convenient” app. Beneficial user experience features are tagged onto “personalisation” features which enable data gathering via the back door, a lengthy privacy policy. Is this a fair trade?
  • Could Meta bring this to the UK? Or can GDPR be enforced by strengthening on-device privacy?

3 responses from the Newspage community

Copy all

Star Quote
Copy

Colette Mason
AI Ethics Consultant at Clever Clogs AI
Facebook's news feed has deteriorated through 2025 and 2026, with users complaining their content is increasingly irrelevant with friends replaced by a stream of algorithmic posts and ads. A cynic might think marketing Forum as a move to "avoid the noise" is conveniently timed.

Meta tried a standalone Groups app in 2014 and killed it in 2017. This version arrives weeks after Reddit posted $625 million in Q1 2026 ad revenue, a market Meta has never cracked.

Analysts forecast Meta's free cash flow to drop 80% in 2026, from £32 billion to £6.3 billion. The company has issued £41 billion in debt and halted share buybacks. It's spending up to £274 million a day on AI infrastructure, while launching an app positioned to capture the most personal data its users will ever reveal.

Is this a cash cow built on people baring their souls 'anonymously', with nothing between their most personal disclosures and one of the most profitable advertising data extraction machines ever built?
Copy

Editor
Mitali Deypurkaystha
AI Governance Strategist & CMO at AISensa
Stalking offences in England and Wales have surged past 135,000 a year, according to House of Commons Library data, yet Meta is launching an app where people may discuss addiction, illness, grief or abuse while handing over precise location and sensitive personal data at the same time. Forum sells the comforting illusion of anonymity, but a single breach could turn pseudonymous support groups into a searchable directory of vulnerable people linked to real identities and real-world locations. That is what makes this different from an ordinary social media privacy scare because the people using these groups are often at their most emotionally exposed. Safe spaces stop being safe very quickly when the business model depends on collecting exactly the kind of information people thought they were hiding behind a nickname.
Copy

Editor
Katrina Young
Futurist | Chief Technology Officer CTO/CDO Ethical AI & Digital Transformation Strategist at KYC Digital
Meta’s Forum app sits at a structural contradiction. The product promises pseudonymous safety in support groups while its App Store privacy label declares collection of health data, precise location, browsing history and identifiers linked to identity, including for advertising-related purposes. The framing of “anonymity” here is doing work the architecture cannot support. Admins can see identities behind pseudonyms. An AI Ask feature aggregates answers across groups. Users experience a nickname; the system experiences a fully attributable person. This is where consent starts breaking down operationally. People disclose differently when they believe they are unidentifiable. When disclosure and identification sit inside the same environment, the safety users perceive may be social, not technical. The real issue is not whether a breach happens later. It is whether users were ever meaningfully anonymous in the first place.