Reddit goes to war against bots: prove that you are human!

Reddit goes to war against bots: prove that you are human!

As AI-generated content becomes more common, Reddit is tightening its stance. The platform announced that it may, in rare circumstances, ask accounts exhibiting suspicious behavior to “verify their humanity.”

The goal is not to turn Reddit into a site with permanent identity control, but to safeguard what still holds value: real people engaging with other real individuals.

Steve Huffman emphasizes this point, stating that these verifications will be exceptional and will not affect the majority of users.

Reddit: Targeted verification, not widespread control

Reddit is not implementing a systematic verification system across the site. Instead, the approach is much more targeted: when an account displays signs of automation or unusual behavior, the platform may trigger an additional step to confirm that a human is indeed behind the screen. Reddit characterizes this as a “rare” measure explicitly aimed at “dubious” accounts—that is, those that arouse suspicion.

This nuance is important. Reddit seems to understand that sweeping measures could clash with its historical culture of relative anonymity. The company is therefore seeking a compromise: to verify the existence of a human without making this verification a widespread obligation to post or read content.

Passkeys, Face ID, World ID: Reddit explores various tools

To conduct these checks, Reddit mentions third-party solutions, such as Apple’s, Google’s, or YubiKey’s passkeys, along with biometric systems like Face ID. The platform also references World ID, a system supported by Tools for Humanity and associated with Sam Altman, as well as, in certain countries or jurisdictions, the use of government-issued ID when local laws regarding age verification mandate it.

However, Reddit asserts its desire to maintain a separation between a user’s real identity and their Reddit account. In other words, the platform seeks to establish proof of human presence, not a complete transition to a visible civil identity.

As always, it will depend on how these tools are integrated into the user experience.

Authorized bots will need to be clearly labeled

In parallel, Reddit is standardizing the labeling of permitted automated accounts. Approved bots will now carry a [APP] label so it is immediately clear that a user is interacting with software rather than a person. TechCrunch explains that developers will need to register their tools to obtain this labeling, fostering a clearer transparency on the platform.

This point is nearly as significant as the verification process itself. Reddit is not waging war on all automation. Instead, the platform is drawing a line between legitimate tools that are properly identified and fake accounts intended to manipulate, spam, or drown out human discussions with synthetic content.

A massive ongoing battle against fake accounts

Reddit also stated that it continues to eliminate harmful bots on a large scale. TechCrunch reports that the company is already removing about 100,000 accounts daily, while improving its reporting tools to help users more easily flag suspicious behaviors. Thus, the aim is not to replace existing moderation with a simple verification process, but rather to add an additional layer against malicious automation.

This figure also highlights the scale of the issue. For Reddit, the question is no longer theoretical: the platform now considers that the proliferation of bots and synthetic content directly threatens the perceived quality of its communities. Huffman has defended the idea for months that Reddit must remain “the most human place on the internet.”

What this means for the Reddit experience

For most users, perhaps nothing, at least in the short term. Reddit asserts that these verifications will not be deployed as a mandatory toll at the entrance. However, the platform is sending a very clear signal: in the future, it will be less concerned about whether a text was assisted by AI than about knowing who is actually behind it.

In summary, Reddit is not banning posts created with AI assistance. The company mainly aims to prevent an internet filled with conversational bots from masquerading as an authentic human crowd. And at a time when the line between assistance, automation, and manipulation becomes increasingly blurred, this is likely one of the most fundamental defenses a social platform can still attempt to maintain.


Scroll to Top