Redact-ID.com

Service Ended

Redact-ID, a privacy-respecting age verification solution, has shut down.

We have decided to end the service due to waning interest from webmasters, end users, and even ourselves.

Service Shutdown Details

Effective immediately, new website registrations have been disabled. For previously registered websites using the service, we would normally keep the service running for a period of transition time, however, there are zero websites still using the service. Thus we are shutting it down immediately.

While we already took care to only collect minimal personally identifiable information, the database has now been fully cleared. What little user / webmaster / site data had been collected is now deleted.

If you wish to link to this notice, please use the wayback machine link, as this server will eventually be deleted.

Postmortem Analysis

Trial Data

While our solution did provide a privacy-preserving method of age verification, the fundamental issue we ran into is that end users simply refuse to undergo age verification -- even when they are offered a solution that can assure them of guaranteed privacy. (Our solution's unique feature was that it provided the ability to verify partially covered IDs, wherein the user could be assured their name/address would not be transmitted.) But even easier than submitting a redacted ID, it was simpler for a user to just visit another non-compliant website which does not require age verification at all, of which there will always be plenty to choose from.

The below data is from one of the adult sites that integrated Redact-ID (they also integrated other contemporary age verification options, providing end users a choice of verification method). This website tracked what users did when presented with a verification demand. The site trialed age verification for a little less than 90 days in certain territories in an effort to comply with foreign legislation in good faith. However, the results were disasterous for this website.

On a sample size of 31,043 unique users (as measured by cookies on their site):

Though these numbers are dramatic, they are consistent with other industry published numbers where AV was attempted.

Following the failure, the site removed Redact-ID and other age verification walls. (It does continue to respect parental controls via the RTA Label meta tag.)

Over time, we found the Redact-ID service failed to get any significant adoption, likely due to similar experiences by other sites. Our service became irrelevant, prompting us to eventually abandon the service and concede there are inherit logical flaws in age verification.

Privacy isn't the (only) problem with AV

Much has been already said elsewhere about the myriad of privacy problems with age verification. Between breaches of other id scanning solutions and others outright having intelligence agency backdoors embedded in them, the privacy problems are well established and taint the industry with a poor reputation.

When creating this service, we thought that if we could solve the privacy problems, we might be able to find a happy midpoint between regulators and webmasters. A service affordable enough to be adopted even by small sites, but private enough to satisfy even the most suspicious of Digital ID skeptics. But the real problem we faced was not solving privacy, but was fundamentally being a product forced on users who did not want it.

Site-based age verification will never be viable for this reason. The people requesting age verification products aren't the same people as the intended consumers of age verification products -- in fact, they have somewhat conflicting motivations. The 'supply' side of the AV equation is driven by legislators and supposed parental activists (in reality, funded by big tech) who are perfectly happy to make it difficult for adults to access adult content, as they are unlikely to use these products themselves. The costs of them imposing a draconian system are effectively externalized onto others. But the 'consumption' side of this equation (adult site visitors) simply wants to find the easiest/fastest/safest path to the content they want to enjoy. That path of least resistance will always be non-compliant sites. You fundamentally cannot engineer or legislate your way around this core conflict.

Legislators are already moving to block non-compliant sites at the ISP level (such as in Canada's prosposed S-209 legislation), but it is hard to believe they will somehow succeed in this endeavour to block all adult content. If even China cannot manage it with their Great Firewall, it does not seem like other nations will have much better success. (And, of course, there are significant free speech issues with this approach.)

The path forward

For the above practical reasons, it becomes clear site-based verification (such as we tried to provide) is not the path forward.

Some technical minded people are pushing zero-knowledge-proofs as a cure-all. However, ZKP has the same underlying problems but worse. The ease-of-use factor is not solved since it requires consumers to get onboarded into a verification ecosystem instead of more readily just go to a non-compliant website. And the privacy is not solved for numerous reasons, whether that be the identity document upload to an opaque server backend, or a build process which creates client builds that may differ from the published open source reference specifications. (Either of these are perfect vectors for intelligence agencies, who as the Persona leak above showed, are very much already at the AV table.)

Will device-based age attestation be a more practical path for legislators? It comes at a great and unnecessary cost to privacy. Social media is already flooded with reports of users being forced to let Google scan their IDs just to open the McDonald's app in Texas or undergo Apple age verification to download any apps in the UK or even visit a web page for developer tools. Google's history as an advertising broker and prior missteps with personal data does not build faith in them.

Lately, much ado is being made about age inference systems. These require collecting and feeding behavior data into AI systems, providing even less privacy to users as every click and page view is analyzed. The accuracy of these system so far have been extremely dubious. It also has the added benefit of being nearly impossible for small-scale sites (both adult and social media/community sites) to meaningfully implement, which is perfect for regulatory capture. So this, too, is a failure in the making.

All the while, parental controls have been providing a comprehensive blocking solution without any privacy violation, but no one is spending money to lobby for that so legislators seem determined to ignore or deride this already proven solution.

What we do know is that site-based verification has failed. The sites that have tried verification walls seem to be moving on to other solutions and strategies, and it is time for us to move on as well. Thank you to those who signed up to try our service out, as well as to everyone who has supported the goal of maintaining privacy on the web.

Redact-ID.com
September 22, 2026