Megalopolis Trailer Included Fake AI-Generated Quotes Attributed to Film Critics

Lionsgate pulled the trailer for "Megalopolis" after it was discovered to contain fake quotes from well-known film critics, generated by AI. The quotes falsely criticized Francis Ford Coppola's previous films. The incident was attributed to an error in vetting by marketing consultant Eddie Egan. Lionsgate has since parted ways with Egan and apologized to Coppola and the critics affected by the fabricated content.

Risk classification

• Primary risk domain: 3 Misinformation
• Primary risk subdomain: 3.1 False or misleading information

The AI system generated false and fabricated quotes attributed to real critics, which were then distributed to the public in a movie trailer.

Additional risk subdomains

• 5.1 Overreliance and unsafe use: The marketing consultant over-relied on ChatGPT for factual research, failing to verify the generated quotes before publication.

Causal factors

• Entity: AI
• Intent: Unintentional
• Timing: Post-deployment

The incident was triggered by ChatGPT generating fabricated quotes (hallucinations) which was an unintentional outcome of the consultant's query.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: Limited Risk. The system involved is a chatbot (ChatGPT) used to generate content, which falls under limited risk applications requiring transparency about AI-generated content.

AI system and alleged parties

• AI system: ChatGPT (OpenAI)
• AI purpose: Writing Assistant; Question Answering
• Behaviour type: Assistant
• Alleged developer: Unknown chatbot developer
• Alleged deployer: Lionsgate, Eddie Egan
• Alleged harmed parties: Pauline Kael, Lionsgate, Francis Ford Coppola, Film critics, American Zoetrope

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Minor
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Negligible, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Financial

Reported: The report does not explicitly state a specific dollar amount of financial loss, but notes that Lionsgate recalled the trailer and parted ways with the consultant.

Indirectly caused: The marketing consultant lost their employment contract with Lionsgate, resulting in personal financial loss.

Inferred additional harm: Lionsgate likely incurred significant unquantified financial losses from the costs of producing, distributing, and subsequently recalling the trailer, as well as potential negative impacts on the film's marketing campaign.

Epistemic

Reported: Yes, the report explicitly describes the creation and dissemination of fabricated quotes falsely attributed to real film critics.

Directly caused: The AI generated fake quotes (such as Kael calling 'The Godfather' 'diminished by its artsiness') which were published in a widely distributed movie trailer.

Indirectly caused: The public was exposed to false historical information regarding the critical reception of classic films.

People affected

• Occurrences reported: 1
• People reportedly harmed: 4
• People reportedly exposed: 10000

Potential causes

Management

• Inadequate Quality Control: Lionsgate management did not enforce strict verification on promotional assets.
• Poor Vendor Oversight: Management trusted the independent consultant without auditing their methods.

Technology

• AI Hallucination: ChatGPT generated plausible-sounding but fabricated critic quotes.
• Lack of Source Verification: The AI model did not verify generated quotes against actual historical texts.

Data Inputs

• Biased Prompting: Prompting for negative reviews forced the AI to fabricate negative quotes.

Human Factors

• Overreliance on AI Output: The consultant trusted AI-generated quotes without verifying original sources.
• Negligent Fact-Checking: The consultant failed to verify quotes before incorporating them in the trailer.

Process and Methods

• Inadequate Vetting Process: Lionsgate lacked a rigorous workflow to fact-check promotional materials.
• Absence of Verification Protocol: No process was in place to match quotes against actual published reviews.

Regulatory Environment

• Lack of AI Content Standards: No industry regulations governed the use of AI-generated content in marketing.

Information quality

• Classification confidence: High
• Reason for confidence: The report provides clear, direct evidence of the AI's role, the specific fabricated outputs, the studio's response, and independent verification of the AI's behavior by Variety.

A marketing consultant used ChatGPT to generate fake critic quotes for a movie trailer, resulting in a commercial recall. This incident represents a commercial vetting failure and AI hallucination issue with negligible national security implications.

• Overall national security impact: Negligible
• Scope: Single nation
• Primary target: No clear primary
• Alleged perpetrator: Marketing consultant

Threat characteristics

• Imminence: Long-term. The incident is resolved and does not represent an active or imminent national security threat.
• Autonomy: Human-controlled. The AI was used as a writing assistant, and human operators made the final decision to publish the content without proper verification.
• Novelty: Established threat. AI hallucinations and human failure to verify LLM-generated facts are well-established phenomena that have occurred in numerous industries.

Impact by dimension

• Physical security: Negligible. The incident is a commercial movie marketing error with no physical security or critical infrastructure implications.
• Information security: Negligible. While involving AI-generated misinformation, the context is entirely commercial movie marketing, with no connection to state actors or national security information warfare.
• Sovereignty: Negligible. The incident is restricted to the entertainment industry and has no impact on state sovereignty or government functions.
• Economic security: Negligible. Financial impacts are limited to private entities (Lionsgate and a consultant) and do not threaten strategic industries or national economic security.
• Societal stability: Negligible. The fabrication of film critic quotes does not threaten social cohesion, civil liberties, or broader societal stability.