Wikipedia talk:Temporary account IP viewer

Is requesting the permission at WP:PERM/TAIV mandatory, or could one also make a permission request at an administrator's user talk page instead? If mandatory, then the sentence "Requests for the permission may be made at Wikipedia:Requests for permissions/Temporary account IP viewer." should then be replaced with "Requests for the permission must be made at Wikipedia:Requests for permissions/Temporary account IP viewer.". GTrang (talk) 04:00, 12 August 2025 (UTC)

The usual practice on Wikipedia is to allow administrators to grant most permissions at their discretion regardless of WP:PERM, but TAIV is special in that the WMF has laid out minimum requirements that cannot be violated (foundation:Policy:Wikimedia Access to Temporary Account IP Addresses Policy#Other users), one of which is that a user must Submit an access request to local administrators in order to be granted TAIV. I suppose it is possible to interpret this as requiring a request be made in a location where multiple administrators regularly review (as otherwise it could have said submit an access request to a local administrator instead). Maybe it might be worth seeking clarification from the WMF on this question. Mz7 (talk) 09:28, 12 August 2025 (UTC)

@SGrabarczuk (WMF): is this a question you are able to answer? (I see you are away on vacation; hope that went well! This is not urgent; take your time.) Best, HouseBlaster (talk • he/they) 22:32, 23 August 2025 (UTC)

Hey, thanks for the ping, and yeah, I'm back! In the global policy, we meant:

• to say that there must be a human request, and the right is not granted automatically
• not to determine whether it's a dedicated RfP page, or a village pump, the main page talk page, or something else - this depends on the size and customs of the local community.

So it's up to you to specify where the requests should (or must, or must not) be made. SGrabarczuk (WMF) (talk) 18:44, 28 August 2025 (UTC)

Thank you!

I've taken a stab at clarifying the policy to note that requests at other venues are permitted but discouraged, and that admins can encourage qualified editors to apply for the permission (e.g. SPI clerks, regulars at AIV). Best, HouseBlaster (talk • he/they) 19:04, 28 August 2025 (UTC)

For someone who edits more often on a phone, can this right be used in the mobile web interface? - Arcrev1 (talk) 10:46, 23 August 2025 (UTC)

Deployment on enwiki has been pushed back to the week of October 6 per Phabricator. I updated the page to reflect this. SuperPianoMan9167 (talk) 20:55, 1 September 2025 (UTC)

do you have to be 18 or older to be a temporary account IP viewer? I am asking this because it functions similar to CheckUser, which requires the user to be 18 or older to get the user right. AquilatorG10   Talk   Contribs  17:02, 29 September 2025 (UTC)

To be a checkuser, it is required to sign foundation:Policy:Wikimedia Foundation Access to Nonpublic Personal Data Policy, which sets the minimum age. This requirement is not needed for this right. – robertsky (talk) 17:58, 29 September 2025 (UTC)

I thought the ANPDP allowed folks under 18 under some circumstances. So this idea that a checkuser needs to be 18 may not be correct. –Novem Linguae (talk) 21:48, 29 September 2025 (UTC)

For reasons that I do not understand (even after extensive discussions about it with WMF Legal), there are two different flavors of the ANPDP. The CU/OS/Arbcom flavor says 18. The VRTS flavor says 16. And, sadly, the TAIV flavor does not exist. RoySmith (talk) 16:09, 11 October 2025 (UTC)

I've been reading various policy docs (always a dangerous thing if you don't want to go down a rabbit hole) and I'm confused about how admins get TAIV access. Foundation:Policy:Wikimedia Access to Temporary Account IP Addresses Policy says Local Administrators ... may opt-in through Special:Preferences on the local project ... these users may choose not to opt-in; access will then not be granted. But here we say Admins will automatically (emphasis in the original) see temporary account IP information which seems at odds with the global policy.

As far as I can tell, the distinction is whether the group you're in has checkuser-temporary-account-no-preference (which is granted to CUs) or just checkuser-temporary-account which is granted to admins. So I think the description here needs to be clarified. RoySmith (talk) 15:50, 11 October 2025 (UTC)

I have updated the wording to address the reality of it all. Best, Barkeep49 (talk) 17:06, 11 October 2025 (UTC)

"Stewards are authorized to make exceptions to the Foundation requirements (criteria 1–4) if someone needs access for a purpose that cannot be reasonably addressed by users who already have this access." Is this correct? Stewards can make an exception to requirement #4???? That doesn't seem like a great idea. --B (talk) 13:30, 2 November 2025 (UTC)

See foundation:Policy:Wikimedia_Access_to_Temporary_Account_IP_Addresses_Policy#Exceptions_to_minimum_requirements. Kovcszaln6 (talk) 13:33, 2 November 2025 (UTC)

The warning said, "Please be aware that editing with multiple temporary accounts may be against the rules of this wiki and you could be blocked if this happens frequently.", however people who clear browsing history don't get to see that warning so maybe we shall show that warning on account creation instead of account exit. ~2025-31341-01 (talk) 08:31, 5 November 2025 (UTC)

@Tamzin@Izno, I've reworded the wording added in . I admit the previous wording (which I wrote) was likely overly strict, but your wording seems overly broad. We seem to be in a legal gray area that WMF Legal has been notified of, but for now I've reworded again (sorry for the accusatory edit summary). Toadspike [Talk] 10:03, 7 November 2025 (UTC)

@Toadspike: The question of to what extent we can name a country or ISP is a good question for Legal, and I'm fine being ambiguous on it for now (although in that case the section should say it's ambiguous, not assume an answer). However, your current wording directly contradicts the policy this is meant to summarize, which says a user may "give information such as same network/not same network or similar" without imposing or implying any restrictions on that. There is no reason for us to say that is "not recommended" when the WMF's position is that it's completely fine. The catch-all "Avoid disclosing more information than necessary" should be enough. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 11:57, 7 November 2025 (UTC)

That's fair, I've removed it. Toadspike [Talk] 12:34, 7 November 2025 (UTC)

Thanks. I've also added a bullet point that just states frankly what we don't know:

The wording of the WMF policy is unclear as to whether disclosure of specific IP-related information (e.g. "~{2025-12345-67 is in France on Orange") falls under the same strict rules as disclosing specific IPs, the less strict rules for describing that information in relative terms, or something else. As of 7 November, the matter is pending clarification from WMF Legal.

Hopefully people can use their best judgment in the interim. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 12:43, 7 November 2025 (UTC)

@MMoss (WMF): Just want to confirm per our discussion on Discord that if I amend TAIVDISCLOSE as follows to resolve the ambiguity here, that is an accurate representation of global policy? This would be a sub-bullet to the TA-to-TA bullet:

The same is generally true of metadata about an IP address, such as its ISP, whether it appears to be a proxy,^[a] or a general location, unless this information is precise enough that it could plausibly be used to identify the user to a greater degree than they have self-disclosed. Note that some ISPs are very small or even belong to a single person, and that some countries have very small populations or have very few English Wikipedia editors. Avoid disclosing more information than necessary; when feasible, favor relative terms like "~2025-12345-67 and ~2025-12345-68 are on the same ISP and geolocate to the same area".

-- Tamzin^{[cetacean needed]} (they|xe|🤷) 22:22, 12 November 2025 (UTC)

Yes, this is an accurate representation. Thanks for working through these nuances! MMoss (WMF) (talk) 23:05, 12 November 2025 (UTC)

TAIV is granted to almost every user in good standing requesting for it. Then this whole exercise of shrouding IP addresses' behind temporary accounts seems pretty pointless. Unless of course, the intention is to gradually make TAIV really hard to obtain and only for select users. ~2025-34073-96 (talk) 18:50, 16 November 2025 (UTC)

Previously anyone, logged in or not, could see where an IP editor was geographically located/ISP used. There have been some valid concerns and issues that have occurred with this. Now only 1,111+ users (at the time of this comment) on Enwiki are able to view this information of the temporary editors. I think this is a notable change in privacy and safety for temporary editors. While there is a high rate of requests being accepted, you also need to consider the editors who are requesting this have been around for a while, and are demonstrating they meet the WP:TAIVGRANT requirements (notably account for six months and 300 edits to English Wikipedia as some evidence of trustworthy behavior. Sure, there might be some added restrictions later on, and some bad eggs that will have this permission revoked, but this is not as pointless as you claim. Zinnober9 (talk) 19:31, 16 November 2025 (UTC)

WP:TA mentions protecting anonymous users from territories with animosity against Wikipedia.

Wouldn't applicants for the TAIV role from democratic nations be favored at large over applicants from undemocratic nations, as part of proving that they're not associated with any governments or organizations known to actively hunt down IP editors? Wouldn't applicants naturally attract more suspicion if they hail from certain countries with governments that treat Wikipedia with said animosity or have totalitarian tendencies? I'm sorry if this sounds snarky or cynical, but it seems sort of logical for certain applicants to have a higher risk factor if they reside in nations that could potentially pressure them into funneling IP addresses to them in an unwarranted manner. ~2025-36592-35 (talk) 00:27, 27 November 2025 (UTC)

In practice, administrators do not profile TAIV applicants by country. That'd probably go against our values of egalitarianism and the right to edit pseudonymously.

The benefit protecting anonymous users from territories with animosity against Wikipedia does exist, in my opinion, but it takes the flavor "before everyone could view your IP address, now only a couple hundred editors with a track record of doing good work for Wikipedia can view your IP address".

If a temporary account is from a country where they could suffer consequences for their editing, the best move is to register an account. An temporary account can be checkuser'd by a couple hundred people (TAIVs), and the logs are unlikely to be scrutinized. A full account can be checkuser'd by a couple dozen highly trusted people (WP:CHECKUSERs), and there is a paper trail of all checks (e.g. WP:SPI cases and checkuser logs) that is paid attention to pretty closely.

Hope that helps. –Novem Linguae (talk) 04:29, 27 November 2025 (UTC)

That'd probably go against our values of egalitarianism: That's not always true; for example, you can't gain CheckUser perms if you're from China, see . All sysops who lived in Saudi Arabia were WMF banned in 2022 (see here). And if you're a sysop on the Swahili Wikipedia, you are automatically topic banned from the GENSEX area (U4C decision). Children Will Listen (🐄 talk, 🫘 contribs) 18:58, 20 December 2025 (UTC)

As an aside, the exact quote from WP:TA at the time read ...helps editors avoid political persecution in territories with animus toward the English Wikipedia. It was only many months later an editor (me) caught this weird focus on only English Wikipedia, and the relevant sentence is now much more in line with what has been quoted here. CapnZapp (talk) 09:32, 26 May 2026 (UTC)

Could Temporary Account IP Viewers have the ability to use auto-reveal feature and why not? ~2025-37397-24 (talk) 18:29, 30 November 2025 (UTC)

You can easily force-enable it by hooking some JavaScript if you aren't an admin, but I don't think you're supposed to do that, and it'll flood the access logs since you'll still fail the server-sided autoreveal check. As for why it's only available to admins, you have to ask Trust and Safety for that. @SGrabarczuk (WMF)? Children Will Listen (🐄 talk, 🫘 contribs) 19:01, 20 December 2025 (UTC)

Hey, thanks for the question and the ping. I'm answering only now because I'm just back from the Polish Christmas-New Year-Epiphany break.

It's an arbitrary choice, as both the feature and the TAIV group itself are in the "exceptions and tradeoffs" area. In principle, the number of users with access to other people's IP addresses and the number and duration of the addresses being shown should be kept minimal. We created the TAIV right to make it available to some non-admins, and we created the auto-reveal feature to make it easier to perform the patrolling tasks for those who need it the most. Making the feature available to TAIVs would be too much of a concession.

(As a fun fact, temporary accounts were built by a team whose previous name was Trust and Safety Product, now it's Product Safety and Integrity, and it's not really part of Trust and Safety.)

Thanks, SGrabarczuk (WMF) (talk) 19:57, 7 January 2026 (UTC)

I asked this over on the Foundation site and haven't gotten a response. I am thinking of asking for TAIV rights, but I wanted to clear one thing up. Would it be acceptable to view a TA IP if a TA is making disruptive edits that may hint at long-term abuse? Even if it is not linked to a particular LTA case. The main scenario that comes to mind is a person making disruptive edits on at least one account with a short editing history, but referring to policies and processes that a new editor would be unlikely to know about. In such an instance might revealing an IP at AIV or SPI be acceptable? I have seen IPs reported at AIV if there is disruption from multiple TAs. TornadoLGS (talk) 04:34, 12 December 2025 (UTC)

@TornadoLGS See the next section, this language has been clarified. Zinnober9 (talk) 16:03, 19 December 2025 (UTC)

Yes, and I've expanded it a bit more, even, to address Tornado's question more directly: [The WMF's policy] is broad enough that it may be better expressed in terms of what it does not include: viewing IPs for no reason at all, out of mere curiosity, or to investigate someone who is engaged in entirely unobjectionable editing. (This is an addition to the general restrictions on out-right bad-faith use of the tool.) -- Tamzin^{[cetacean needed]} (they|xe|🤷) 15:06, 20 December 2025 (UTC)

Just noting that I've updated TAIVDISCLOSE to incorporate the changes made in foundation:Special:Diff/541536. The main change here is broadening the consecutive-block exception to all admin actions, which was something I'd reported to Legal as a gray area after I unblocked an IP after a TA on the IP complained about the block.

The other big change in that foundationwiki diff, which doesn't come across here because it isn't about TAIVDISCLOSE, is a further weakening of the criteria for using TAIV in the first place. The sentence There must be a valid reason to investigate a temporary user was removed, leaving us just with the very short list of prohibited reasons to use the tool for political control, to apply pressure on editors, or as a threat against another editor in a content dispute. Furthermore, discovery of inappropriate behavior was added as a valid reason to use the tool, which clarifies that speculative use of TAIV is okay as long as you have some reason to think it might find a policy violation. I think this is further evidence for the interpretation that the WMF does not intend to regulate use of the tool for all but the most obviously inappropriate reasons, and people should not feel worried that they're going to get TAIV yanked because they looked at an IP that had made a maybe-constructive-maybe-not edit.

There's also now an explanation of how the policy interacts with auto-reveal, including that Use of the IP auto-reveal mode by users who regularly use temporary account IP addresses in the course of investigations and enforcement is permissible, as long as any automatically revealed IP addresses are used solely for the purposes described above. Probably this policy should have a brief section on auto-reveal. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 08:42, 19 December 2025 (UTC)

Actually, here, I've updated the rest as well. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 08:49, 19 December 2025 (UTC)

Yeah, I’d agree with this change. ~2025-42676-23 (talk) 09:54, 26 December 2025 (UTC)

This help request has been answered. If you need more help, you can ask another question on your talk page, contact the responding user(s) directly on their user talk page, or consider visiting the Teahouse.

Is there a reason on why is using the tool, logged for each use? ~2025-43890-78 (talk) 10:26, 30 December 2025 (UTC)

The Wikimedia Foundation logs it for legal compliance reasons. It could also potentially be used to track down who leaked TAIV info, but that isn't its primary purpose. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 12:35, 30 December 2025 (UTC)

MMoss (WMF), to add to this discussion, are we allowed to say things like ~2026-12345-67 and ~2026-12345-68 are on proxies, The TA is on a proxy, The following group of TAs are on proxies? (without being specific about which proxy they're using or the geolocation of their proxies, of course) Some1 (talk) 13:35, 21 March 2026 (UTC)

Yes! Thank you for helping clarify that point. MMoss (WMF) (talk) 21:15, 25 March 2026 (UTC)

Thank you, I have added this to WP:TAIVDISCLOSE: Special:Diff/1345404919. Some1 (talk) 22:55, 25 March 2026 (UTC)

@Some1, to clarify that a bit, you're allowed to disclose whether an editor is likely to be on proxies, but first you'd have to be pretty sure that they are. The TAIV tool is usually telling you that a proxy has been spotted on that IP recently. That's not the same thing as "this editor is on proxies". Most of the time, when the TAIV tool reports proxies, there is no reason to believe the editor themselves is on one. -- asilvering (talk) 00:52, 5 April 2026 (UTC)

Also, the permissibility of disclosing proxy information is already established in the first sentence of that bullet point (along with the footnote that such disclosure is separately, if mildly, limited by a different policy). -- Tamzin^{[cetacean needed]} (they|xe|🤷) 10:54, 5 April 2026 (UTC)

It used to be my perception that there used to be a high level of scrutiny placed on who exactly was entrusted with checkuser abilities. We also know the WMF has had occasional problems where tools were used in pursuit of unwanted attention.

On some wikis - I have not checked what scrutiny is given here - the TAIV bit is very easily obtained, and while a small group of users are then tasked with viewing the secret logs of these these hardly-vetted TAIV users, these are the exact persons (CheckUsers) who were already busy with a considerable workload.

The concern I have is that the wording in this document suggests that the TAIV users get access to the cookie-tracked IP information previously reserved for CheckUsers, and that this access to several IP addresses would allow TAIV rights holders to narrow down the location and commuting patterns of a potential victim.

This is different from the previously "available" procedure of using tools to try to guess whether IP addresses that have appeared in the same page history belong to the same individual, especially as some may not even edit the same page twice.

To illustrate this point a little, we are now dealing with over 400 TAIV rights holders on the English Wikipedia, and 46 CheckUsers supposedly patrolling their logs. That does not seem like a feasible task.

Those with knowledge of the matter, please comment.

~2026-24738-83 (talk) 11:29, 7 May 2026 (UTC)

I think this problem is minimal. Temporary accounts are close to anonymity, so one cannot mine information from this unless they know their victim well. For example:

• John edits under the account Example, and I have IP address information linked to it.
• A device uese a temporary account to edit at places A and B.

The latter being what TAIVs can get, I think is much less that what can be used to track someone in real life. -- Least Action (talk) 12:45, 2 June 2026 (UTC)