Gmail’s Inbox Sorting System Reportedly Reduced Visibility of Political Emails and Campaign Calls-to-Action

Google's Gmail inbox sorting algorithm for political emails was reported by presidential candidates, nonprofits, and advocacy groups for having negative impact on call-to-actions, allegedly suppressing donations and impeding political actions.

Risk classification

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

The incident stems from Gmail's classification algorithm failing to robustly and accurately distinguish between commercial marketing and political/advocacy communication, leading to high rates of misclassification.

Additional risk subdomains

• 1.1 Unfair discrimination and misrepresentation: The algorithm treated different political candidates and organizations highly unevenly, resulting in disparate outcomes for their campaigns.

Causal factors

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

The incident was caused by Gmail's deployed classification algorithm, which unintentionally filtered political communication into promotions and spam, impacting democratic engagement.

EU AI Act risk tier

• Risk tier: 4 Minimal or No Risk

Minimal or No Risk: The EU AI Act explicitly lists 'spam filters' as an example of AI systems that pose low or negligible risks to users and society, falling under Risk Level 4.

AI system and alleged parties

• AI system: Gmail inbox sorting algorithm (Google)
• AI purpose: Spam Filtering; Content Curation
• Behaviour type: Autonomous
• Alleged developer: Google
• Alleged deployer: Google
• Alleged harmed parties: political organizations, political candidates

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 Minor, indirect Minor
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Minor, indirect Minor
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Financial

Reported: The report explicitly mentions that the drop in email open rates suppressed donations for political campaigns and advocacy groups, though it does not quantify the exact monetary loss.

Indirectly caused: The reduction in email open rates by approximately 50 percent or more for Gmail subscribers led to an unquantified loss of donations and fundraising revenue for multiple advocacy groups and political campaigns.

Inferred additional harm: Given that political campaigns and nonprofits rely heavily on email for fundraising, the systemic diversion of emails to spam and promotions likely resulted in millions of dollars in lost donations across the affected organizations.

Differential treatment

Reported: The report explicitly describes uneven and differential treatment of political candidates and organizations by the algorithm.

Directly caused: Gmail's algorithm treated senders unevenly, delivering 63 percent of Pete Buttigieg's emails to the primary inbox while delivering 0 percent of Elizabeth Warren's and Joe Biden's emails there.

Indirectly caused: This uneven delivery resulted in disparate visibility and engagement rates for different political campaigns and advocacy groups based on how the algorithm classified them.

Inferred additional harm: It is likely that many other political candidates and advocacy groups not included in the study experienced similar differential treatment, affecting their public reach and campaign viability.

Democracy

Reported: The report explicitly describes concerns from advocacy groups that Gmail's algorithmic curation of political emails impedes the public's ability to engage in political action, which is fundamental to a healthy democracy.

Directly caused: Gmail's algorithm diverted nearly 90 percent of political emails from the primary inbox, suppressing petition signatures and political mobilization.

Indirectly caused: The curation of political communication by a private company's algorithm centralizes control over democratic engagement and fundraising, potentially impacting election outcomes.

Inferred additional harm: The systemic filtering of political speech by major email providers likely weakens democratic participation and public awareness of critical political issues on a national scale.

People affected

• Occurrences reported: 1
• People reportedly harmed: 171
• People reportedly exposed: 171

Potential causes

Management

• Monetization Conflict of Interest: Suggesting buying ads incentivizes poor free reach.
• Prioritizing Organization Over Reach: Focusing on user inbox organization over sender visibility.

Technology

• Algorithmic Sorting Flaws: Algorithm misclassifies political emails as promotions or spam.
• Inconsistent Classification: Similar emails treated differently across candidates and groups.
• Lack of Explicit Primary Label: Primary inbox is determined by excluding other categories.

Data Inputs

• User Interaction Bias: Algorithm relies heavily on aggregate and individual user actions.
• Mislabeled Senders: List sharing led to incorrect attribution of email senders.
• Metadata Classification Limits: Algorithm relies on metadata fields like X-Gmail-Labels.

Human Factors

• Default Settings Reliance: Users rarely disable default tabbed layout settings.
• User Interaction Variance: Varying user habits cause unpredictable sorting behavior.
• Lack of User Training: Users do not manually correct sorting errors often.

Process and Methods

• Opaque Sorting Logic: Lack of transparency on how the algorithm classifies emails.
• Inadequate Edge Case Handling: Official non-campaign emails were sent to promotions.
• Lack of Appeal Mechanisms: Senders have no way to contest incorrect sorting.

Regulatory Environment

• Absence of Platform Regulations: No rules govern algorithmic curation of political speech.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide detailed empirical data from a structured four-month experiment analyzing over 5,000 emails across 171 political groups. The findings are supported by statistical drops in open rates and direct statements from affected organizations and Google representatives.
• Ambiguities identified: The exact proprietary mechanics of Gmail's classification algorithm are not fully disclosed, and the extent to which individual user interactions would override the default sorting remains unquantified.
• Alternative interpretations: The filtering could be interpreted purely as a standard spam/promotions filter working as intended to reduce inbox clutter, rather than a systemic threat to democratic engagement.

Gmail's automated email sorting algorithm unintentionally diverted a large majority of political emails to spam or promotions folders, unevenly affecting political candidates and suppressing fundraising. While raising concerns about the algorithmic curation of democratic discourse, the national security impact is minor and manageable within standard regulatory and platform governance procedures.

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

Threat characteristics

• Imminence: Long-term. The incident represents an ongoing systemic issue regarding algorithmic curation rather than an active, immediate national security crisis.
• Autonomy: Full autonomy. The Gmail sorting algorithm operates and executes classification decisions automatically without real-time human intervention.
• Novelty: Evolved capability. While spam filtering is an established technology, the systemic and uneven impact of algorithmic curation on national political campaigns represents an evolved threat vector.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or kinetic threats were impacted by the email filtering algorithm.
• Information security: Minor. While the algorithm affected the distribution of political information at scale, there is no evidence of a coordinated foreign information warfare or intelligence operation.
• Sovereignty: Minor. The filtering impacted political campaigns and fundraising, which are core to democratic processes, but did not represent a hostile compromise of government functions.
• Economic security: Negligible. The incident caused unquantified fundraising losses for political groups but did not threaten strategic industries or national economic security.
• Societal stability: Minor. The algorithm resulted in uneven treatment of political speech and candidates, but did not cause civil unrest or systematic human rights abuses.