This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
An Israeli-based creator produced and circulated a deepfake video featuring unauthorized AI-generated likenesses of numerous Jewish celebrities, including Scarlett Johansson, protesting antisemitic remarks made by Ye (formerly Kanye West). The video, which used the song 'Hava Nagila' and depicted celebrities wearing shirts with offensive imagery, prompted a public statement from Scarlett Johansson condemning the misuse of AI and calling for urgent government legislation to protect citizens from the dangers of synthetic media.
A deepfake video allegedly depicted Scarlett Johansson, Jerry Seinfeld, David Schwimmer, Lisa Kudrow, Steven Spielberg, Adam Sandler, Sacha Baron Cohen, Natalie Portman, Jake Gyllenhaal, Lenny Kravitz, Adam Levine, Drake, Mark Zuckerberg, Jack Black, Mila Kunis, Ben Stiller, and Woody Allen protesting Kanye West’s antisemitic remarks. The video falsely attributed statements to them, using AI to manipulate their likenesses without consent.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to conduct large-scale disinformation campaigns, malicious surveillance, or targeted and sophisticated automated censorship and propaganda, with the aim of manipulating political processes, public opinion, and behavior.
Also classified as
The deepfake video generated false information by misleading the public into believing the depicted celebrities had participated in the campaign.
The unauthorized replication of celebrity likenesses and identities devalues their personal and professional creative control.
The incident involved the intentional creation and viral dissemination of a deepfake video featuring unauthorized celebrity likenesses to influence public opinion at scale.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was caused by human creators who intentionally used post-deployment generative AI tools to create and distribute the unauthorized deepfake video.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
Israeli digital marketers created an unauthorized deepfake video featuring 15 Jewish celebrities protesting Kanye West's antisemitic remarks. While highlighting growing concerns over synthetic media and celebrity likeness rights, the incident has minor national security implications, representing a cultural dispute rather than a coordinated information warfare campaign.
The reports describe the AI system being used by digital marketers as a tool to generate synthetic video and voice likenesses of celebrities. The AI did not act autonomously or as an assistant, but was directly operated to produce specific outputs.
Limited Risk: The incident involves AI-generated content such as deepfakes, which are subject to transparency obligations under the Act to ensure users are aware they are interacting with or viewing AI-generated media.
The primary function of the AI system in this incident was to generate deepfake video and voice likenesses of celebrities without their consent.
The reports provide consistent, detailed accounts of the deepfake video, its creators, the celebrities depicted, and the public reaction, particularly from Scarlett Johansson. The facts are well-documented across multiple reputable news sources.
None
Technology
Accessible Deepfake Tools
Generative AI tools make it easy to replicate realistic human likenesses.
Lack of Technical Guardrails
AI systems lack built-in blocks to prevent unauthorized celebrity generation.
Data Inputs
Unauthorized Likeness Data
Celebrity photos and videos used as training data without explicit consent.
Human Factors
Deliberate Non-Consensual Use
Creators intentionally used celebrity faces for an activist campaign.
Public Vulnerability to Fakes
Audiences struggle to distinguish realistic AI deepfakes from reality.
Process and Methods
Reactive Platform Moderation
Social networks rely on post-upload labeling rather than proactive blocking.
Lack of Consent Verification
No process exists to verify celebrity consent before publishing AI content.
Management
Weak Corporate Ethics Oversight
Creative agencies fail to restrict employees from using non-consensual AI.
Prioritizing Outreach Over Safety
Creators focused on campaign impact rather than ethical AI guidelines.
Regulatory Environment
Paralyzed Legislative Action
The US government has failed to pass federal laws regulating AI deepfakes.
Weak International Standards
Countries like the US and UK declined to sign international AI declarations.
Vetoed AI Safety Bills
State-level attempts to regulate AI safety have been vetoed or delayed.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
