Social media algorithms out LGBTQ+ identities before self-disclosure, study finds

On July 6, 2026, new research from Newcastle University reveals that social media algorithms are increasingly identifying users’ sexual orientations or gender identities prior to their conscious self-disclosure. The study, published in Gender, Place & Culture, introduces the concept of "algorithmic outing," where platforms infer a user’s LGBTQ+ identity based on engagement patterns.

Understanding Algorithmic Outing

Lead author Dr. Justin Ellis highlights how users are often presented with LGBTQ+ content before they have identified as queer themselves. Participants reported that their feeds, particularly on platforms like TikTok, were filled with LGBTQ+ material based on their interactions, such as likes and follows. "Several participants told us the algorithm 'knew' they were queer before they did," said Ellis.

This phenomenon raises significant concerns regarding privacy and safety. While it can provide a space for exploration, it also poses risks when such identities are revealed in public or semi-public environments.

Hybrid Spaces and Personal Safety

The research further investigates how LGBTQ+ individuals navigate hybrid spaces—areas where online and offline lives intersect. Common locations include cafes, public transport, and workplaces, where algorithmic outings often occur. Ellis notes, "Seeing queer content appear on your screen while on a bus or in a cafe isn't a neutral experience. Screens are visible, and algorithms don't account for risk, context or personal readiness."

Participants described employing various strategies to manage their online presence, such as using privacy settings, multiple accounts, or passive browsing to avoid unwanted attention or harassment. This vigilance reflects a growing awareness of how algorithms can impact their safety.

Calls for Greater Responsibility from Platforms

The study calls for increased transparency and responsibility from digital platforms regarding their algorithms. Participants suggested practical design improvements, including clearer consent mechanisms and privacy-by-design features. Such tools could help users quickly exit sensitive content in unsafe situations.

Ellis emphasizes the importance of accountability, stating, "If platforms are going to identify who their users are, they also need to take responsibility for how—and where—that identity becomes visible to ensure community safety." This call to action highlights the need for a more responsible approach to algorithm design that prioritizes user safety.