Artificial intelligence can infer personal characteristics from publicly available online content with measurable accuracy. Research published during 2025 and 2026 demonstrates that large language models (LLMs) can identify personality traits, estimate behavioral tendencies, and generate psychological profiles from digital footprints including social media posts, professional profiles, and written narratives.
A practical starting point for understanding publicly available online information is domain registration lookup, which provides access to domain registration data when that information is publicly available or not protected by privacy services.
AI Uses Public Digital Footprints to Infer Personality
Studies have shown that AI systems can analyze multiple forms of online content to estimate the Big Five personality traits:
- Openness
- Conscientiousness
- Extraversion
- Agreeableness
- Neuroticism
The prediction process relies on language patterns, vocabulary, writing style, posting behavior, and publicly shared information rather than direct psychological testing.
A 2026 study published in Global Business and Organizational Excellence analyzed online content from 348 individuals. Researchers compared AI-generated personality predictions with results from the IPIP 50-item personality assessment. The study concluded that publicly available online content can approximate traditional psychometric survey results and reflects measurable aspects of real-life personality.
Social Media Posts Provide Predictive Signals
Researchers have tested whether AI can infer personality from long-term social media activity.
A study published in the Journal of Personality analyzed two years of Facebook posts from 1,214 Italian users. Gemini 1.5 Pro and GPT-4o generated Big Five personality estimates that were compared with participant self-reports.
The study reported several measurable findings:
- Cross-model agreement improved when predictions from multiple AI models were combined.
- Test-retest stability increased when personality estimates were aggregated across multiple time periods.
- Correlations with self-reported personality reached:
- 0.31 for Openness
- 0.27 for Extraversion
- 0.24 for Agreeableness
- 0.23 for Conscientiousness
- 0.18 for Neuroticism
The researchers concluded that digital footprints contain sufficient behavioral information for AI systems to generate statistically meaningful personality estimates, although prediction accuracy remains moderate rather than perfect.
LinkedIn Profiles Also Reveal Personality Patterns
Professional networking profiles have also become subjects of AI personality research.
A study published in the Journal of Business and Psychology evaluated 406 LinkedIn profiles using Microsoft Copilot powered by GPT-4.
Researchers measured both reliability and validity.
The findings included:
- Good test-retest reliability across repeated analyses.
- Moderate correlations with verified personality assessments.
- Intelligence, openness, and extraversion produced the strongest prediction accuracy.
- AI relied on consistent profile characteristics when generating personality estimates.
The study also identified systematic limitations, including positivity bias and differences in prediction accuracy across demographic groups.
Narrative Writing Produces Strong Personality Signals
Research published in Nature Human Behaviour found that LLMs can estimate personality traits using brief written narratives.
The study demonstrated that AI-generated ratings:
- Converged with participant self-reports.
- Predicted daily behavior.
- Predicted mental health indicators.
- Outperformed traditional language-processing methods used for the same task.
The authors concluded that modern LLMs provide scalable personality assessment using ordinary written language rather than structured psychological questionnaires.
Online Conversations Also Contain Predictive Information
Researchers have extended personality inference beyond social media.
An analysis of 62,090 ChatGPT conversations involving 668 participants evaluated whether AI could infer personality traits from conversation history.
The researchers found:
- Prediction accuracy exceeded random classification for multiple personality traits.
- Relationship discussions produced stronger prediction performance than general conversations.
- Personal reflection conversations generated the highest inference accuracy for extraversion.
The authors identified personality inference from AI conversations as a measurable privacy consideration because conversational data contains psychologically relevant information.
AI Personality Prediction Has Measurable Limitations
Current research consistently reports that AI personality inference is informative but imperfect.
Documented limitations include:
- Moderate rather than high correlations with validated psychological assessments.
- Reduced prediction accuracy for some personality traits.
- Bias toward average personality scores.
- Performance differences across AI models.
- Demographic bias identified in some datasets.
Researchers recommend combining multiple observations across different time periods to improve reliability.
Online Transparency Has Increased Importance
The growing ability of AI to infer personal characteristics has increased interest in digital transparency and reputation management.
Organizations and individuals increasingly evaluate how publicly available information may contribute to AI-generated profiles, particularly when information originates from multiple online sources.
Additional discussion of this topic is available in this analysis of online reputation in the AI era.
Research Supports AI-Based Personality Inference
Recent peer-reviewed research demonstrates several consistent findings:
- AI can estimate personality traits from publicly available online content.
- Social media activity contains measurable psychological signals.
- Professional profiles provide sufficient information for partial personality assessment.
- Brief narrative writing enables accurate prediction of several personality characteristics.
- Long-term digital footprints improve prediction stability.
- AI-generated personality estimates remain probabilistic rather than definitive and should not replace validated psychological assessment.
Current evidence indicates that AI systems can reveal meaningful aspects of an individual’s online identity using publicly available digital information. The strongest research supports personality inference as a statistically measurable capability rather than a speculative application, while also documenting important limitations in accuracy, bias, and generalizability.
