Baltimore Lawsuit Alleged DraftKings and FanDuel Used Machine-Learning-Driven Targeting to Exploit Vulnerable Gamblers

In Baltimore, the city sued DraftKings and FanDuel, alleging that the companies used predictive modeling with machine-learning algorithms, extensive user data, personalized promotions, push notifications, and VIP programs to identify and exploit vulnerable bettors, including people showing signs of gambling disorder. The complaint alleges these systems were designed to maximize user lifetime value and contributed to financial and public-health harms among Baltimore residents.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The platforms utilize predictive algorithms to conduct targeted manipulation of vulnerable users' behavior for the companies' financial gain.

Additional risk subdomains

• 5.1 Overreliance and unsafe use: The algorithmic targeting fosters severe behavioral and material dependence, leading to compulsive gambling and addiction.

Causal factors

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

The alleged harm is driven by proprietary algorithms deployed post-deployment that intentionally target and exploit vulnerable users to maximize platform engagement.

EU AI Act risk tier

• Risk tier: 1 Unacceptable

Unacceptable Risk. The report describes exploitative systems targeting vulnerabilities based on age or behavioral disorders, and AI used for behavioral manipulation leading to physical, psychological, or financial harm.

AI system and alleged parties

• AI system: predictive modeling with machine-learning algorithms
• AI purpose: Behavioral Modeling; Content Recommendation
• Behaviour type: Autonomous
• Alleged developer: Flutter Entertainment, DraftKings
• Alleged deployer: FanDuel, DraftKings
• Alleged harmed parties: People with gambling disorders, Baltimore residents

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. 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 Minor, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Substantial, indirect Minor
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Differential treatment

Reported: Yes, the report describes algorithms that identify, target, and exploit the most vulnerable users.

Directly caused: The proprietary algorithms treat users differently by identifying those with gambling disorders and targeting them with aggressive inducements.

Inferred additional harm: Systematic exploitation of vulnerable demographic groups, particularly young males aged 18 to 24, who are disproportionately affected.

Psychological

Reported: Yes, the report explicitly describes gambling-related harms, compulsive gambling, and gambling disorders.

Directly caused: The algorithms target vulnerable users to induce compulsive betting, directly contributing to gambling addiction.

Indirectly caused: The families of addicted users experience severe distress, as evidenced by helpline calls from parents of young men.

Inferred additional harm: Widespread psychological distress, anxiety, and clinical addiction are highly likely among thousands of targeted users in Baltimore.

People affected

• Occurrences reported: 1
• People reportedly harmed: 50
• People reportedly exposed: 1000

Potential causes

Management

• Prioritizing LTV Over Safety: Management focuses on maximizing lifetime total value of vulnerable users.
• Deceptive Marketing Strategy: Designing platforms specifically to hook and exploit users for profit.
• Exploitative VIP Programs: Using VIP programs to personalize and encourage continuous betting.

Technology

• Predatory Targeting Algorithms: Algorithms altered to target users likely to suffer from gambling harm.
• Push Notification Algorithms: Proprietary algorithms optimize notification timing to induce next bet.
• Exploitable App Logic: Bots mimic disordered gamblers by opening apps in middle of the night.

Data Inputs

• Exploitation of User Data: Using troves of data to identify and target vulnerable user behaviors.
• Personalized VIP Profiling: Data on user habits used to personalize inducements and maximize LTV.

Human Factors

• Compulsive Mobile Betting: Young men unable to restrain impulses due to constant phone access.
• Illusion of Low Risk: Bonus bets lead users to believe gambling is not risky.
• Disordered Gambling Behavior: Platforms designed to create and exploit disordered gambling habits.

Process and Methods

• Relentless Push Notifications: Sending constant push notifications with calls to action to bet often.
• Predatory Bonusing Practices: New user promotions forcing multiple bets within seven days.
• Lack of Financial Checks: No robust financial checks or VIP program restrictions in the US.

Regulatory Environment

• Weak US Gambling Regulations: US lacks the financial checks and VIP restrictions found in the UK.
• Inadequate Consumer Protection: Existing laws failed to prevent abusive trade practices by operators.

Information quality

• Classification confidence: High
• Reason for confidence: The legal allegations and the mechanisms of algorithmic targeting are clearly described in the report, allowing for a highly confident assessment of the risks and harms.
• Ambiguities identified: The specific technical names and exact operational parameters of the proprietary algorithms are not disclosed.
• Alternative interpretations: The defense may argue that the algorithms are standard marketing optimization tools rather than predatory systems.

The City of Baltimore's lawsuit against sports betting platforms highlights the domestic regulatory and social challenges of commercial machine-learning systems. While the predatory targeting of vulnerable individuals causes localized social and financial harm, the incident lacks the scale, intent, or foreign state involvement to pose a significant threat to US national security.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: DraftKings and FanDuel

Threat characteristics

• Imminence: Long-term. This represents an ongoing commercial and regulatory concern rather than an active national security crisis requiring urgent intervention.
• Autonomy: Full autonomy. The machine-learning systems autonomously analyze user data, personalize promotions, and dispatch push notifications without real-time human intervention.
• Novelty: Evolved capability. Represents a highly sophisticated evolution of predictive marketing algorithms specifically optimized to exploit behavioral vulnerabilities and addiction.

Impact by dimension

• Physical security: Negligible. No physical threat, kinetic capability, or critical infrastructure compromise is associated with this commercial gambling software incident.
• Information security: Negligible. The incident involves commercial marketing algorithms and does not compromise national intelligence, classified models, or state information operations.
• Sovereignty: Negligible. This is a local consumer protection lawsuit against private corporations, posing no threat to state authority, territorial control, or core federal government operations.
• Economic security: Negligible. While causing individual financial harm, the incident does not threaten national financial stability, strategic industries, or the US technological competitive advantage.
• Societal stability: Minor. Predatory algorithms targeting vulnerable populations for commercial gain cause localized social and financial distress, but this remains manageable within standard legal and regulatory frameworks.