Gannett Halts AI-Generated High School Sports Articles After Series of Errors and Public Backlash

Gannett, a newspaper chain, temporarily halted its AI experiment that used a tool called LedeAI to generate high school sports articles. The decision came after several articles produced by the AI showed glaring errors, repetitive language, and awkward phrasing, drawing criticism and mockery on social media.

Risk classification

• Primary risk domain: 7 AI system safety, failures, & limitations
• Primary risk subdomain: 7.3 Lack of capability or robustness

The LedeAI system failed to perform reliably, generating articles with coding errors, placeholder text, and repetitive phrasing, demonstrating a lack of capability and robustness in real-world deployment.

Additional risk subdomains

• 3.1 False or misleading information: The AI system generated and published sports articles containing factual errors and incomplete information to public readers.
• 6.3 Economic and cultural devaluation of human effort: The deployment of automated sports writing tools occurred alongside newsroom layoffs, reflecting the displacement and devaluation of human journalistic work.

Causal factors

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

The incident was caused by the LedeAI system generating faulty articles after being deployed by Gannett, which was an unexpected and unintentional outcome of the automation process.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The system is used for generating text content, which falls under transparency obligations for AI-generated content to ensure users are informed of its nature.

AI system and alleged parties

• AI system: LedeAI
• AI purpose: Writing Assistant; Automated Content Curation
• Behaviour type: Tool
• Alleged developer: LedeAI
• Alleged deployer: Gannett
• Alleged harmed parties: Student athletes, Newspapers relying on Gannett, General public, Gannett

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 Negligible
• 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

Epistemic

Reported: The report explicitly describes epistemic harm in the form of erroneous, repetitive, and incomplete sports articles published across multiple local newspapers.

Directly caused: Several local newspapers published sports dispatches containing errors, placeholder text, and repetitive phrasing, misleading readers about high school sports events.

Indirectly caused: The publication of low-quality automated content led to public mockery and a temporary erosion of trust in the journalistic standards of the affected newspapers.

Inferred additional harm: It is likely that thousands of local sports readers were exposed to inaccurate or poorly formatted information before the articles were corrected or updated.

People affected

• Occurrences reported: 1
• People reportedly exposed: 10000

Potential causes

Management

• Cost-Cutting Job Reductions: Laying off 6 percent of staff led to reliance on unvetted AI tools.
• Inadequate Vendor Evaluation: Management failed to properly vet LedeAI before deploying it widely.

Technology

• Coding and Programming Errors: Bugs in the AI software led to placeholder text and repetitive phrasing.
• Flawed Text Generation: The AI produced repetitive language and awkward phrasing in sports stories.

Data Inputs

• Missing Mascot Data: Lack of team mascot data left placeholder text in published articles.
• Lack of Sports Knowledge Base: The AI lacked detailed contextual data to write comprehensive sports stories.

Human Factors

• Lack of Human Copyediting: Editors published AI articles without reviewing them for errors first.

Process and Methods

• Inadequate Quality Control: No pre-publication review process was in place to catch AI errors.
• Premature Automation Deployment: The tool was deployed across multiple markets before thorough testing.

Information quality

• Classification confidence: High
• Reason for confidence: The reports clearly describe the nature of the AI system, the specific errors it made, the newspapers affected, and the organizational response. There is little ambiguity regarding what occurred, and the classifications are well-supported by the text.
• Alternative interpretations: The incident could be viewed primarily as an economic story about newsroom automation and layoffs, but the immediate cause of the pause was the technical failure of the AI system.

An automated sports-writing tool used by Gannett published articles with repetitive phrasing and placeholder errors, leading to public mockery and a temporary pause of the tool. The incident has negligible national security implications, representing a commercial quality control failure rather than a threat to national infrastructure, sovereignty, or security.

• Overall national security impact: Negligible
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. Represents a minor quality assurance and automation issue in commercial media rather than an active national security crisis.
• Autonomy: Full autonomy. The AI system generated and published articles directly to live public news sites without human editorial review.
• Novelty: Established threat. Software and template-generation errors in automated journalism are established issues with existing precedents.

Impact by dimension

• Physical security: Negligible. No physical security or critical infrastructure systems were affected by this incident.
• Information security: Negligible. The incident involved localized, automated sports reporting errors with no connection to systematic disinformation or intelligence compromise.
• Sovereignty: Negligible. The incident was limited to local commercial journalism and did not impact government functions or sovereignty.
• Economic security: Negligible. No strategic technologies, financial systems, or critical supply chains were compromised; impact was limited to minor commercial reputational damage.
• Societal stability: Negligible. No systemic human rights violations or large-scale threats to social cohesion occurred; public response was limited to social media mockery.