X Teases Private Dislike Button for Replies to Curb Spam

X is preparing to introduce a dislike button for replies, signaling a fresh attempt to fine-tune conversation quality and curb spam on the platform owned by Elon Musk. The feature was teased by head of product Nikita Bier, who shared a screen recording that revealed a thumbs-down control appearing alongside the familiar like and bookmark icons.

Early glimpses suggest the button won’t function as a public scoreboard. Instead, it feeds private signals into X’s ranking systems to better understand what users don’t want to see. That design choice echoes lessons from other platforms that have tried to reduce pile-ons while still capturing negative feedback.

How the New Reply Dislike Button Will Function on X

In the demo shared by Bier, tapping the thumbs-down triggers a “Reply feedback” prompt with options such as Not interested in this post, Incorrect or misleading, AI-generated, Spam, and Report post. X emphasizes the feedback is private, with no public dislike counter on display.

That privacy matters. Public downvotes can incentivize brigading and humiliation, while private signals help train ranking models to demote low-quality replies and spam. The addition of an “AI-generated” option is notable, hinting at a feedback loop aimed at synthetic or low-effort content that increasingly clutters replies.

Bier also hinted that the company will target the economics behind spam, saying the financial incentive to post junk replies will drop and could even turn negative. Combined with a dislike signal, that suggests X will link user feedback to enforcement and monetization controls for reply spammers.

Why X Is Pushing a Private Dislike Option for Replies Now

Replies are the platform’s pressure point. That’s where engagement-bait, crypto scams, and AI spam congregate, especially under high-visibility posts. A low-friction, private feedback channel could let everyday users help the algorithm push irrelevant or misleading replies out of view without publicly shaming individual accounts.

X has leaned into community-driven moderation tools before. Community Notes (formerly Birdwatch) is a prominent example; the company has pointed to meaningful reductions in engagements with posts once a credible note appears. Private negative signals from a dislike button would complement that approach by acting at the reply level and at greater scale.

What History and Prior Research Suggest About Downvotes

Twitter, prior to its rebrand as X, experimented with downvotes on replies and reported that testers frequently used the control to flag off-topic or unhelpful responses. Crucially, those downvotes were not public, and internal findings indicated they helped surface higher-quality replies without triggering dogpiles.

Other platforms offer cautionary context. YouTube removed public dislike counts and later said it saw fewer coordinated “dislike attacks” on creators. Reddit and Stack Overflow use downvotes to maintain relevance and quality, but they also contend with brigading when communities mobilize. The balance, research shows, lies in treating negative feedback as a ranking signal rather than a public verdict.

There’s also the misinformation angle. MIT researchers found that false stories travel faster than truthful ones on social networks and were 70% more likely to be reshared. Private, scalable feedback about low-quality or misleading replies could give ranking systems a valuable counterweight, especially when users select “Incorrect or misleading.”

Risks, Misuse Potential, and Key Design Trade-Offs for X

Any dislike mechanism can be misused. Coordinated groups may still mass-flag viewpoints they dislike, potentially suppressing marginalized voices. Pew Research Center has reported that around 40% of U.S. adults have faced online harassment, and critics worry that negative signals can amplify that pressure if not calibrated carefully.

Keeping the metric private mitigates some harm, but implementation details matter: how signals are weighted, whether fresh accounts can impact rankings, how often feedback must be corroborated, and what role human review plays when users choose “Report post.” Transparent guidelines and auditing—internal or third-party—will be key to trust.

What This Change Means for Everyday Users and Creators

For everyday users, the pitch is simple: a quick way to tell X “less of this” when a reply misses the mark. For creators and brands, reply quality could improve if spam is downranked faster, though some may worry about silent penalties when their content is misinterpreted or brigaded.

The “AI-generated” option is a notable signal for the era of automated engagement. If X ties that feedback to stricter detection and lower visibility—or reduced revenue opportunities—for low-quality synthetic replies, creators who rely on genuine conversation may see cleaner threads and better signal-to-noise.

The Bottom Line on X’s Private Dislike for Replies

A dislike button for replies, designed as private feedback rather than a public scoreboard, is a pragmatic move for X. It builds on past experiments, borrows lessons from rival platforms, and targets the reply swamp that undermines user trust. The impact will hinge on tuning and transparency—how X weights signals, resists abuse, and links feedback to enforcement. If the execution matches the intent, replies on X could finally become more useful than noisy.