/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Meta submits data to the EU showing that pop up content warnings stopped Facebook users sharing 25% of flagged posts and 38% on Instagram; TikTok reports 29%

Bloomberg :

Bloomberg

Context & Ripple Effects

The EU's transparency regime has long measured moderation by removals — its annual report on platform takedowns showed flagged-material removal rates slipping from 71% in late 2019 to 62.5% by spring 2021. This submission reframes that debate around a different metric: whether flagged posts get shared at all.

The timing matters for Meta specifically. Since its January policy changes, advertisers have raised brand safety concerns fearing more harmful content, while Meta argues its lighter-touch approach works — it separately claims to have halved US content removal mistakes. The EU data is Meta's first hard evidence that warnings suppress spread without taking posts down.

First-order effects

  • Meta can now answer advertiser brand-safety objections with efficacy data: Instagram warnings stopped sharing of 38% of flagged posts and Facebook 25%, meaning harmful material reaches fewer feeds even when it stays up.
  • TikTok's comparable 29% figure gives the EU a cross-platform baseline for judging whether warnings are a real intervention or a fig leaf.

Second-order effects

  • Rival platforms face pressure to publish equivalent warning-efficacy numbers to the EU, since TikTok already has — abstaining starts to look like hiding weaker results.
  • If warnings satisfy regulators, Meta gains cover to lean further on non-removal interventions, reinforcing the direction of its January moderation overhaul rather than reversing it.

Third-order effects

  • EU oversight is drifting from counting what platforms delete toward measuring what actually spreads — a structural shift that could make share-suppression rates, not takedown rates, the standing compliance benchmark under future enforcement.
  • For the industry, 'friction over removal' becomes a monetizable moderation posture: fewer wrongful takedowns, defensible safety claims, and less exposure to both regulator fines and advertiser boycotts.

The trend: Content moderation is shifting from takedown-volume metrics toward measured suppression of harmful spread, with EU submissions turning warning efficacy into the new comparative yardstick across platforms.

Discussion

  • @lukolejnik Lukasz Olejnik on x
    Platforms removed accounts/content spreading disinformation about the Russian war in Ukraine. TikTok: 1,704 accounts, Twitter: 75,000. Google: 8,000,000 ads blocked, 60 State-funded sites removed, 9k videos from YouTube, 9k channels. #DigitalServicesAct https://twitter.com/...
  • @aggichristiane @aggichristiane on x
    TIL only 25% of Facebook users in Europe won't share something after it's flagged for having been fact-checked Disinfo reports for EC's Code of Practice are out today https://disinfocode.eu/... https://twitter.com/...
  • @deutschjill Jillian Deutsch on x
    Well er this doesn't age well already https://disinfocode.eu/... https://twitter.com/...
  • @lukolejnik Lukasz Olejnik on x
    Platforms released first #DigitalServicesAct disinformation code of conduct compliance. Huge, lengthy documents. Very difficult to navigate. Someone designing this process has no idea about clarity of presenting data. What a shame. #digitaleuambassador https://disinfocode.eu/... …
  • @stworg Stephan Lewandowsky on x
    Google's looks professional and it is clear that they engaged with the task (one can still debate whether they did enough or whether it's 100% accurate). For example look at this table: 2/n https://twitter.com/...
  • @stworg Stephan Lewandowsky on x
    Someone did their homework. Now look at the equivalent section in Twitter' report: 3/n https://twitter.com/...
  • @stworg Stephan Lewandowsky on x
    Twitter fails blatantly and comprehensively and reprehensibly to meet its European obligations: https://www.politico.eu/.... The reports are in the public domain here: https://disinfocode.eu/... It is informative to compare Twitter's to Google's for example. Some obervations 1/n
  • @mattnavarra Matt Navarra on x
    Social Media Warnings Rarely Stop Users Sharing Potentially False or Misleading Posts The % of cases where social media users in the EU continued to share a post after being shown a warning that the content was “unverified”: FB: 75% TikTok: 71% IG: 62% https://www.bloomberg.com/.…