What is GARM (Global Alliance for Responsible Media)

Is GARM still active for brand suitability?

No. The WFA said on 9 August 2024 that it was discontinuing GARM activities, citing allegations that had "significantly drained its resources and finances" (WFA statement). The announcement came just days after X launched an antitrust lawsuit against the WFA, GARM and its members (The Drum). The WFA statement names no successor, and as of 2026-10-13 we found no body that took over maintenance of the framework. What remains in use is the framework itself, applied by platforms and vendors under their own policies.

What GARM was

GARM was a voluntary cross-industry initiative the WFA created in 2019, set up after the Christchurch mosque shootings were livestreamed on Facebook (WFA statement). It brought together advertisers, agencies, platforms and trade groups. Its main output was the Brand Safety Floor and Suitability Framework, published in 2020, which gave the industry shared definitions for eleven content categories, including adult and explicit sexual content, arms and ammunition, hate speech, terrorism, and death, injury or military conflict (Media Rights Agenda, 2020-10-08). The WFA describes the toolset as the Brand Safety Floor and the Adjacency Standards Framework, which brand owners used to build their own brand-specific frameworks (WFA statement).

How the floor and the risk tiers work

• The floor is content no advertiser should support. Hive's documentation for its GARM brand suitability API defines floor risk as "content not suitable for any advertiser" (Hive docs)
• Above the floor, each category is graded by risk. Hive classifies four tiers (low, medium, high and floor), with high and medium risk suitable only for some advertisers depending on risk type and tolerance (Hive docs)
• IAB Tech Lab's human-labeled benchmark of GARM labels uses the 11 GARM categories and the values minimal, low, medium, high and floor; it was last updated in 2022, before the shutdown (IAB Tech Lab Brand-Suitability-Test-Benchmarks)
• Each advertiser chooses which tiers to exclude in each category, according to its own risk tolerance

How suitability tiers are set today

• Platforms set their own tiers. TikTok, for example, offers Expanded, Standard and Limited inventory, chosen by the advertiser according to brand identity, goals and risk tolerance (TikTok for Business)
• Verification and classification vendors keep GARM's category and tier vocabulary in their products, as in Hive's GARM brand suitability classes (Hive docs)
• Since no body maintains the framework, any change to a category definition is made by each platform or vendor on its own. Ask your vendor which version of the categories its labels follow and when it last changed them
• Commentary after the shutdown argued that ending GARM does not by itself bring the industry closer to better brand safety (Adweek opinion)

How Mixpeek relates to GARM

Mixpeek offers taxonomies as a primitive: documents are classified against a taxonomy through retrievers, with confidence scores, and video is broken into scenes with extractors for frames, speech transcripts and on-screen text. A team can load GARM-style categories and tiers as a custom taxonomy and classify content at ingest, then look the stored labels up when a placement decision is made. Mixpeek also ships IAB Content Taxonomy 3.1. See brand safety and brand safety vs brand suitability for video.