A political ad-transparency library is a searchable public database — Meta's Ad Library is the model most other platforms have copied — that discloses who paid for a political or social-issue ad, roughly how much they spent, and who the ad reached. On October 24, 2025, the European Commission preliminarily found that Meta's and TikTok's versions of this tool still leave outside researchers with, in the Commission's words, only partial or unreliable data, a finding made under the EU's Digital Services Act.
The finding matters beyond Brussels because these libraries are the primary public evidence base for how political advertising actually circulates online. When the underlying tool is unreliable, every downstream study — on election spending, on disinformation, on which audiences see which political messages — inherits that unreliability.
What does a political ad-transparency library actually disclose?
Meta's Ad Library, described on Meta's own Transparency Center, is billed as a “comprehensive, searchable database for ads transparency” covering Facebook and Instagram. For political, electoral, and social-issue advertising specifically, Meta discloses additional detail beyond what ordinary commercial ads get: spend ranges, estimated reach, and the identity of the funding entity behind the ad. Meta says these sensitive-category ads stay visible in the library for seven years, whether the ad is still running or not; ads targeted at audiences in the EU get an added layer, staying visible while active and then archived for a year after they stop running.
A separate tool, the Ad Library API, is meant for a narrower audience: vetted researchers get deeper access, including an ad-targeting dataset that shows the criteria advertisers selected — age, location, interest categories — for political, electoral, and social-issue ads dating back to August 2020, across more than 120 countries. That distinction matters: what the public sees in the consumer-facing library is a fraction of what the API is supposed to hand researchers who study the system at scale.
The gap between the two tiers matters because the public library and the vetted-researcher API answer different questions. A campaign-finance reporter checking who paid for a single ad can usually get that from the public search box. A researcher trying to establish which demographic groups a political campaign targeted across thousands of ads, or whether a platform under-served certain audiences with fact-checks, needs the targeting-criteria dataset that only the API exposes — the same layer the Commission's finding singles out as the one where access has been burdensome and the resulting data partial or unreliable.
What did the European Commission find wrong with Meta's and TikTok's versions?
The Commission's preliminary finding, announced on its Digital Services Act enforcement page, concluded that both Meta and TikTok are in breach of their obligation to give vetted researchers adequate access to public platform data. The Commission characterized the platforms' data-access procedures as burdensome and said researchers who go through them often end up with data that is partial or unreliable — data the Commission said is needed to study, among other things, users' exposure to illegal or harmful content involving minors.
Meta faces two additional preliminary findings that are not about the ad library specifically: that Facebook and Instagram impose unnecessary steps before a user can report illegal content, using design choices the Commission called manipulative, and that the platforms' appeals process does not let EU users fully explain or submit evidence when they contest a moderation decision, according to TechCrunch's reporting on the Commission's announcement.
Henna Virkkunen, the Commission's executive vice-president overseeing digital policy, framed the stakes in institutional rather than punitive terms: “Our democracies depend on trust. That means platforms must empower users, respect their rights, and open their systems to scrutiny,” she said, according to the Commission's own announcement of the findings.
How did the two companies respond?
TikTok told TechCrunch it has made substantial investments in data-sharing and that nearly 1,000 research teams already have some form of access to its data. Meta said it has made recent compliance improvements and that it continues to negotiate with regulators on the outstanding issues, per the same report. Neither company's response, as reported, disputes that the Commission opened the finding — both frame it as a disagreement over whether existing access is sufficient, not over whether more access exists to be given.
What happens next?
A preliminary finding is not a final decision. Under the process the Commission described, Meta and TikTok can review the investigation file, contest the findings, and offer to fix the identified gaps before the Commission issues a formal ruling. If the Commission ultimately confirms a breach, the Digital Services Act allows fines of up to 6% of a company's global annual revenue, according to TechCrunch's account of the Commission's statement. No decision on fines had been issued as of the Commission's October 2025 announcement, and the matter remains open.
Why this is a mechanism problem, not just a compliance one
The design flaw the Commission is pointing at is structural: a transparency library is only as useful as the completeness of the data behind its search box, and neither the public interface nor the researcher API discloses how complete that underlying data actually is. A journalist or academic pulling numbers from Meta's Ad Library has no independent way to know whether an advertiser's spend range is accurate or whether the archive is missing ads that ran and were removed before indexing. The Commission's preliminary finding is, in effect, an assertion that this gap is large enough to call the access obligation unmet — not evidence, on its own, that any specific number in the library is wrong, since the Commission has not published an audit of individual line items alongside the finding.