AI Policy

Meta Oversight Board orders removal of two AI deepfake videos

Meta Oversight Board orders removal of two AI deepfake videos Image: Primary
Meta's Oversight Board on 17 September ordered the company to remove two AI-generated Facebook videos and said Meta's deepfake rules are "consistently and fundamentally inadequate." The decisions are binding on Meta. One video, posted in November 2025, used a Scottish Labour councillor's likeness to make comments about refugees she did not make; it drew more than 5,000 views with an album photo. Meta had found no rule breach and no need for an AI label. A board majority said the clip violated hateful-conduct rules and should have carried a "High Risk AI" label. A second case involved a young Muslim campaigner; Meta automatically closed the first report and appeal. The board issued recommendations including broader high-risk labels, demotion or demonetization, and a wider definition of unwanted manipulated imagery. Meta has 60 days to respond.
Sources
The Next Web
In this story
Meta
Published by Tech & Business, a media brand covering technology and business. This story was sourced from The Next Web and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
AI
AI

Moonshot AI's Kimi K3 now available on Amazon Bedrock

AWS has made Kimi K3 from Moonshot AI available on Amazon Bedrock, describing it as the first open-weight model to reach 2.8 trillion parameters. The model combines native vision capabilities with a 1-million-token context window...

AWS Machine Learning Blog, AWS
AI Infrastructure
AI Infrastructure

AWS launches GPU-aware SageMaker HyperPod Inference Gateway

AWS announced general availability of the SageMaker HyperPod Inference Gateway, a Kubernetes-native routing addon that places inference requests using real-time GPU signals such as KV cache utilization, queue depth, LoRA adapter r...

AWS Machine Learning Blog
Policy AI
Policy AI

Ninth Circuit rejects expanded DMCA theory in GitHub AI case

The Ninth Circuit rejected an attempt to use DMCA Section 1202 as a broader source of copyright liability in litigation involving code used to train large language models, according to EFF. The court held that the absence of copyr...

Deeplinks
AI Policy
AI Policy

OpenAI sets framework for disclosing model-misalignment incidents

OpenAI announced a framework for publicly disclosing AI misalignment incidents and released details of several examples identified during the past year. Under the framework, employees can report incidents to senior safety and ali...

WIRED