Judge Dismissed Roblox AI Training Class Action
The court ruled that the plaintiff failed to prove Roblox intentionally removed copyright-management tags from assets.
Updated on Oct. 5, 2026 in Artificial Intelligence
Live Poll
Should companies be legally required to honor 'NoAI' tags on artist content used for training?
A federal judge in San Jose has dismissed a class action lawsuit against Roblox brought by artist Austin Beaulier. The case alleged that the platform stripped "NoAI" metadata from 3D models to train artificial intelligence models.
Why it matters
The ruling hinges on the interpretation of content management information under the Digital Millennium Copyright Act, setting a precedent for how tech platforms manage data ingestion for AI training. This decision underscores the legal challenges creators face when proving the intent behind automated data processing.
The case focused on the Objaverse-XL dataset, which houses millions of 3D works. The plaintiff alleged that machine-learning preprocessing pipelines converted these 3D assets into numerical inputs while simultaneously stripping content management tags.
The players
Roblox
A global gaming platform and engine provider that maintains an extensive virtual asset ecosystem.
Austin Beaulier
The artist and plaintiff who initiated litigation regarding the unauthorized use of 3D models in AI training.
Beth Labson Freeman
A U.S. District Judge in the Northern District of California overseeing civil litigation involving copyright and tech infrastructure.
The details
The lawsuit claimed that Roblox ignored 'NoAI' tags, which serve as metadata indicating that a digital work is off-limits for AI training. Under the Digital Millennium Copyright Act, such tags are considered content management information, which identifies the work and its terms of use. The judge ruled that the plaintiff did not provide sufficient evidence that Roblox intentionally bypassed these markers during the conversion of raw assets into machine-learning-ready inputs.
Timeline
March 2026: Austin Beaulier filed the class action lawsuit against Roblox.
October 2, 2026: Judge Beth Labson Freeman issued the ruling to dismiss the case.
The Tech Race
This case highlights the growing friction between massive AI training datasets like Objaverse-XL and individual digital creators. The ruling clarifies the evidentiary burden required to hold platform operators accountable for metadata stripping under the Digital Millennium Copyright Act.
This dismissal limits the immediate legal avenues for individual creators to challenge how their 3D assets are utilized in automated machine-learning pipelines. Artists and developers should continue to monitor how platforms implement metadata standards, as this ruling emphasizes the high threshold for proving intentional copyright infringement in AI training.
The takeaway
The court's decision signals that plaintiffs must demonstrate clear intent when alleging that AI training pipelines purposefully bypass copyright protections. Observers should track subsequent rulings regarding the Objaverse-XL dataset to see if further copyright claims emerge that provide stronger evidence of tag removal.
Further reading
For broader analysis on how legal frameworks are adapting to generative modeling, visit Artificial Intelligence.
Source note: This article includes information reported by Court House News Service.
Live Poll
Should companies be legally required to honor 'NoAI' tags on artist content used for training?
