A new academic study examines whether X's June 2024 policy change—making likes private—affected user engagement with reputationally risky content. Researchers analyzed over 154,000 posts and conducted a survey experiment, finding no detectable platform-wide increase in likes for such content. The results suggest hiding like identities produces limited behavioral change.

The study contrasts user intentions with actual behavior. While survey participants reported a modest willingness to like risky content when likes were private, this didn't translate to measurable changes in group-level engagement. The findings indicate engagement may be driven by a narrow set of high-usage or automated accounts, not broad user sentiment.

This research provides a data-driven look at a major social media platform's real-world experiment. It questions whether privacy features can meaningfully alter user behavior at scale. Future work could examine how other engagement signals, such as reposts or replies, are affected by similar visibility changes on X and other networks.