As a Jira user, I do not want to interact with AI, ever, for both ethical and practical reasons.

User control is mandatory. There's a growing number of companies who've been measuring AI usage as a part of performance evaluation, which means needing to ask an admin to create/maintain an opt-out list puts a target on those users' backs. So, the setting not only needs to be per-user, it needs to be private to that user so admins can't e.g. run a query that tells them which users have the AI turned off.

The "Ask AI" pop-up has ruined Confluence/Jira for me.  Until that pop-up (and preferably all AI functionality) is removed, I no longer speak favorably about these tools.  I have gone from a vocal proponent to a held-hostage user who longs to move to an alternative.

Echoing b62e7cc86eac , I also am being driven even crazier by the recent addition of the "Ask AI" pop up every time I try to copy or edit text lately.

For context, part of my job is condensing data from Jira tickets into one, moving reports from one ticket to another. I am copying text from Jira tickets all day long, and this is more than obnoxious - it's like dodging pop ups on a spammy website at my workplace. I should also add that depending on where the text is located that you're copying, the "Ask AI" pop up blocks tools like the icons in the formatting toolbar.

As an admin, I need to be able to restrict AI from specific users on my site, or on specific managed accounts.
We have employees that by virtue of their job mainly handles sensitive or restricted data, so being able to say "No AI for you, buddy" is a must.

As a user I want to be able to disable AI features for my account regardless of site.
I may be "forced" to use Atlassian Cloud products by virtue of my job but have moral, ethical, or principal objections to the use of AI.

is there any target date for this ? Can we take part in Beta ? 

The path that is suggested for the 'fix' - allowing admins to suppress this at the Group level – does not address the need for multiple reasons. Your admin/user customers have been quite clear in the feedback and the change request, and that is for users to be able to control this at the user level. I as an admin don't need the extra (and extraneous) burden of defining Groups and managing group assignments for this purpose. We need this change as requested, at the user level.

If for whatever reason Atlassian wants an additional more aggregated control (emphasis on additional) it would be much more useful to have this at the Project/Space level. Groups are unwieldy and your customers whose admins have to fit in Jira/Confluence admin duties into a piece of their overall role (in my case as a program manager) do not have time, nor desire the added complexity of managing this in a group context.

Apologies if I'm just missing it, but everything I see here is discussing Jira. I'd like to add Confluence to the request. Specifically, I do not have Atlassian Intelligence on for any of my applications (Jira, Confluence), but I see "Ask AI" every time I edit anything in Confluence.

I am with many others that have no interest in using it and want it removed from the assist.

(Honestly, I'm really annoyed that ANYTHING pops up when I go to edit text.)

I want to echo others who have expressed a desire for this to also be available as an individual user setting.

I have colleagues who use AI. I'm not going to try to stop them from using AI. However, I personally avoid using AI as much as possible as I have ethical concerns about generative AI and LLMs. So to me the AI features in Jira are just frustrating clutter.

I want to have a user setting to disable all of these features for myself.

444512db38c5 As others mentioned, the request from the majority of respondents here is to have this be a user-level setting. I would like to be able to turn off the AI features for my account individually, regardless of the wider group's settings

444512db38c5 thank you for the update. I too have some concerns about the verbiage here regarding groups of users vs. users. If only enabled at the group level, this wouldn't address many of our concerns, let alone my personal concerns due to the group my role falls into. We really need this to be at the user level. 

> Thank you for the feedback - We are working on this capability for admins to enable AI for specific group of users.

This is not what we asked for. The request was that users can en- or disable AI in their own Atlassian settings without needing an Admin, given that AI in general is available in the Atlassian account.

This is great 444512db38c5 - really happy that this is coming to fruition. Looking forward to it real soon ...

I do want what this ticket is about - ie. giving org admins the ability to enable it for users or groups only - but as 0eb8a7cf51e9 if this can be added as part of this or as a next phase, that would be brilliant too!

Thank you and Atlassian!

I personally was voting for and watching this ticket to get the capability for Admins to allow or block AI features for specific users or user groups so that we don't have to enable the feature for ALL of our users.

We are working on adopting Atlassian Intelligence and Rovo for our company, and the main blocker here is the fact that when we flip the switch, it enables ALL of the features for ALL users without any way for Admins to limit who has access to these features.

Being able to limit which users or groups have access to these features is almost a necessity at this point for us.

The ability for users to also be able to turn it off as a user setting would be a "nice to have" feature, but not nearly as important, to my company, as the Admin setting which would allow us to rollout this feature to a targeted group prior to expanding the availability to other users who want to opt-in to Atlassian Intelligence and Rovo.

444512db38c5 

Please read this ticket comments again: everyone is requesting user settings to disable AI, not some other admin settings.

Thank you for the feedback - We are working on this capability for admins to enable AI for specific group of users.

This whole post is ignoring the fact that there is a moral justification for enabling users to choose if their data is used in non-parent-company made/focused AI infrastructure.( also ... i had to start work ... so am posting this response unfinished, sorry)

I am sorry, 050d2367638e, but you are effectively asking an employee to leave their position of work over something that could simply be opted out from in account settings. Because asking their organisation to disable something just because they do not want it involved in their work / life / personal data (the latter of which probably holds human rights conversations) which then disables those who DO want to use it will immediately paint said folks as 'problems' especially if they're in a place where upper management are just blindly following the 'must can haz AI' thing.

Luckily I don't have this upper management issue, but I do have that concern about my job, my welfare, and making a living. All this for something that could (and should) just be a toggle in my account settings.

I say this because you're positing a world where-in users are opted in to a system of data analysis that they do not have an option to be included within. That system is markedly NOT for their parent organisation's benefit.

A company WILL NOT give a monkey's about whether or not an individual user is utilising AI interactivity in their interactions within a given product, they certainly would not enforce it.

That being said, they may well encourage its use, especially for those who clearly show a need for it.

We have settings to turn the screen dark ... I think that a button, or automated assessments of our text, can easily be selected as something to flick a switch in our user settings.

It would be good to have an option in account settings to disable AI and Rovo as an individual user.

Collecting data from individual users in AI without their consent might cause legal issues both because of privacy (GDPR) and individual author rights. Also GDPR gives the right to any individual to request deletion of their data, and deleting that data from an AI model might not be easy without deleting the entire AI LM.

I will also check back with our corp legal if the current AI implementation is according to EU law.

It's good that some/all AI things can be turned off at organization level, but convincing an entire organization to adopt a user's request is a challenge. We should at least be able to hide this stuff as a user to reclaim our UI space and stop the constant popups begging us to use them. They are of absolutely no use to me and the area of my screen they claim is growing by the week.

I'm PM for an agency and while some of the team may use the AI, particularly in the issues, I don't like it and would prefer to remove it. I want to read all the details, comments and not a summary. It's just in the way of the work.

It's been a year. It's clear they're not going to address this issue, because being able to say they're "AI first" and impress investors is more important that actually making a product that works. Enshittification is here to stay!

This should have always been an opt-in or easy-to-opt-out feature, as someone who tabs through Jira tickets / uses the keyboar extensively, this is an absolute nightmare for accessibility and also the general user experience sucks as a result.

Plus, like other people have said, I don't want to use it and so it shouldn't eat up as much space as it does.

Atlassian support and I have filed a related ticket, https://jira.atlassian.com/browse/AI-1196 to clarify that even if AI is disabled globally, the "Ask AI" button still appears (and is non-functional). I support removing that button if AI is disabled.

I strongly support, at a minimum, giving us the control to remove it.

As a end user, I should be able to toggle this off and on. Please add this feature, it seems to me to be fairly basic functionality, anyone would expect. 

Also a CSS styling to hide (or dim) all texts generated by AI.

Or don't make this text appear with the name of a human but make it a comment by Jira bot, or something like that.

I don't need AI, I don't want to feel stupid and useless. 

I found out how to remove those underlined word silly definitions.

I also found out how to remove those silly premade comments.

But there are still many other cluttering AI buttons and stuff all around, that I cannot disable.

There should be a single user option to switch everything AI off, at the same time, in all Atlassian cloud.

I work in a big company but all users have different preferences.

Some are hyped by AI while I prefer personal stuff expressed by human beings.

My sentiments echo those expressed by others.  In summary, users need the ability to disable/deactivate AI features if they don't want/need to use them.  

Mandated presence of AI features reduces my satisfaction and encourages me to consider moving to other products.

1. Name of the organization, your role

Company: Candela Technologies, IT Systems Admin  (Tech Support)
Atlassian site:  Site Admin (Jira and Confluence)

1. Why do you need user/group level control? What impact will you have if these controls aren't provided?

No one at our company uses or wants to use or see any popups for any "AI feature" for Jira or Confluence. 
Please either 1) Allow Admin to turn off pop-ups and 'menu items' for the entire site, 2) Have a group called "AI Features" which allows membership for people who want to use/know more AI features (in our case, no membership).  2) Allow users to 'turn off' any mention of AI tools, pop-ups, advertisements, or suggestions.

In short, a) make it an 'App' feature.. opt-in for users who want to explore, b) don't advertize it by cluttering up UI space.

1. How are you using AI today in the absence of these controls?

No.  Please stop the madness, no mention of AI anywhere and anytime !

This should not be an 'Admin' option, this should be a USER option.

Craziness.

The real estate taken up by the Ask AI button on Confluence tables would be much better used by providing me a way to perform literally any table-specific operation on the table. I've never in my life felt the need to "ask AI" anything and the work I do doesn't lend itself to pat answers that can be emitted by simply digesting the internet.

Regarding the generative ML that has been added to Jira comments, all of its suggested comments are things that could be answered by reading the ticket or by looking at any of the data associated with the ticket like "who is working on this?" or "what is the status of this ticket." As a user, I'd like to turn it off so I can remove the clutter from the comments UI, personally. If I actually went down the road the generative system is pushing me I'd sound very obtuse.

And if I'm going to use a generative ML-based system to summarize information I'm going to want that system to cite sources so I can be sure they're real.

I find it incredibly offputting to have AI features forced on users in any software, but having it forced onto me at my job is a bridge too far and is infuriating. I have no use for the AI summaries, I have ignored the feature as long as it's been available in jira, but suddenly every ticket I look at (200+ per month for context - my sole job is working in Jira tickets) has this field auto expanded and I have to X out of it each time. Please give admins the ability to disable AI for individual users. This is inconvenient and distracting, and everyone doesn't have a use case for this. I don't need a summarization of the Jira ticket I personally wrote. Stop this madness. 

1. InMotion Software, Project Manager
2. Some users find AI tools beneficial, while some find them actively distracting and harmful. Allowing more individualized experiences will give more users a better experience.
3. I am not—they are in the way and hamper my ability to use Atlassian tools. The presence of AI that I cannot turn off makes me want to stop using Atlassian and shift to other platforms.

This is the biggest blocker stopping us from utilizing AI in our production environment. We only want to allow users who understand the product's functionality to have access to the AI features. Please introduce a way to granularly control access in Rovo AI and Atlassian AI. 

+1 please

Especially the IA summary. Fisrt, the recap is not generated in the best language, second, it is very annoying to seem typing text while reading a ticket.
And this feature is not usefull for me since I create and update the ticket I work on.

+1 please

This is for our current instance. Would not like to provide organization name publicly. Happy to talk directly if needed. I am 1 of the 3 org admins on our site of +- 500 licensed users.

I would like to enable AI for org admins first and then only for more users/groups before all users in our instance.

We have not enabled and used AI yet because of privacy/compliance reasons.

Thank you

1. RELEX Solutions, System specialist (de facto interim Atlassian system owner)

2. Considering the environmental impact of "AI" (and I'd like to emphasize the air quotes around the "intelligence" part), it's borderline criminal to force "AI"-generated search results on users by default, with no option to disrupt or prevent them.

It prolongs my search process, making for worse UX, because believe it or not: I'm usually able to find what I need in the traditional search results within a few seconds. Loading times were already suboptimal, now they're even worse. As a first- and second-level support agent for Confluence and Jira, I am dealing with complaints from users who are equally impacted and slowed down. "AI" makes us waste working time.

3. Since "AI" results are not reliable and need to be fact-checked and often corrected, I only use "AI" as a last resort if I can't find/figure something out myself. Every once in a blue moon I have to ask "AI", only to realize it gives me no new information than what I had already figured out the "old-fashioned" way, plus 50% of the time it adds some some gratuitous hallucinations that I need to debunk first before ending up in the same place I already was.

1. Org name unnecessary, role is that of a technical director.
2. I need user-level controls because many users wish to disable this, and the admins won't do it, probably because they have higher priorities and don't have time to learn new ways of disabling new, unrequested obstacles.

The impact of not being able to disable AI is that the decontextualized, inaccurate, slow, and confusing content slows down me and my team and is, at best, a significant distraction to confluence search. And, the UXD is not.

1. I am using AI effectively, on an ad hoc basis, because I, like most people, know what I want when I want it.

1. Name of the organization, your role: Org not listing; Executive Officer and Atlassian relationship owner
2. Why do you need user/group level control? What impact will you have if these controls aren't provided?  AI cannot be enabled for advanced ROVO capabilities without having ability to provide privileges to individuals.  Audit guidelines. AI governance mandates.
3. How are you using AI today in the absence of these controls?  We cannnot use Atlassian Rovo intelligence capabilities 

1. Name of the organization, your role
   Company Name not provided. Role: Operations Technician, not an Jira administrator but with high privilges in the software.
2. Why do you need user/group level control? What impact will you have if these controls aren't provided?
   For users that does not want to use the "AI" it adds distracting blinking in the search and unwanted opinions in issues. 
3. How are you using AI today in the absence of these controls?
   I'm not. I'm avoiding it because of the intrusion into my work.

This is obnoxious.  All i want to do is add a comment and now i have to do something first before i can add a comment.  AI is not useful in any of my projects

This "feature" gets in the way. I'm sure some people may find value in it, but I do not.

As many other users have requested, we need to have the ability to turn off the AI feature at the user level. It should be optional for individuals, not a one or all situation.

I've been subscribed to this issue for many months and all I see is more and more people complaining about this... When is this going to be taken care of by the Jira team ? Otherwise everything has been said in other comments...

Please can you prioritize this. Ever since someone at our ~500 user company turned on these AI anti-features our Confluence pages have been overrun with pop-up ads, useless buttons and unsolicited hyperlinks. Even if you block the UI distractions with uBlock Origin, it still hijacks clicks to do random stuff that we never asked for.

It's absurd to force functionality on every user that not only is clearly an optional add-on but also costs a lot of resources to run. Even leaving aside the costs of training these models, the client-side JavaScript is so heavy that all the styling and popups appear several seconds after the page content loaded, making it even more distracting.

Role: senior software engineer who just wants to get my work done, not be harassed by calls to action to engage with functionality I will never use.

I am a capable writer (and typist) and know what I want to communicate.  I find the constant AI prompts to be distracting (and annoying).  For sanity sake, please make them user configurable!  

(And thank you for getting rid of the comment AI which was beyond annoying).

Role: Scrum Master for multiple teams

When scrolling to the Comments section of an Issue in Jira, we're recently (and suddenly) seeing an unwanted prompt to add a comment "powered by Atlassian Intelligence". We NEED a way to turn this off, it is annoying and distracting when often the purpose of reviewing comments is during a work session on a shared screen, and it just gets in the way/delays getting to the info we are actually looking for. 

We don't need prompts for Comments, we're professional adults who know how to manage our work. Please make it possible to turn these off!

I'm a cross-team program manger for a large, multi-department project/team

Martyn Bristow - Tools Admin for ThermoFisher Digital Science

We want to disable Atlassian Intelligence in our regulated documentation spaces to avoid confusion

Blinking, flashing, AI "features" in Confluence are giving me a migraine, please make it stop! Robots are to be made available if needed, not forced on us by some board member to increase profit margins.

As a user, the AI features bring nothing to the UI but distraction.

Not being able to disable UI integration at user level is very surprising and missing.

I do not want any of the AI features. They are very distracting and the blinking colors and popup windows are an active detriment to me using the product.

Please allow me to disable these features as a user. I do not want to use AI.
I can write sentences myself that are clearer and more useful because i have the context.

If you have to beg me to use a feature, its not a good feature.

The AI features delivered to date are a distraction, not a tool.

The UI is already too busy. This makes it worse.

Allow users to enable/disable. Opt-in.

As with ALL AI (and in general most 'new features' at this level) this should be a choice available:

• At admin level for a whole tenancy/instance/whatever
• At a user level for a user to completely opt out of any given AI experience

... and also all the other reasons that others are saying ... too ... but I cannot stress enough that individual users should b allowed to choose to turn such features (and an AI's ability to use their data) on and off themselves.

Additionally, just a little soapboax moment ... folks should always try to ensure that they know that using AI is not the same as RPA.

Nothing but disdain for the AI features. They inflict real harm on an already disorganized and crowded interface.

We just need to disable AI features site-wide.

Please make it a package that can be disabled/enabled like any other feature.

This was done for Confluence, so it should be possible for Jira.

Adding my vote. I've yet to find a useful underlined definition, and they are distracting.

Nothing but disdain for the AI features. They inflict real harm on an already disorganized interface.

Hi, same here. Recently, Jira has started highlighting pieces of ticket descriptions with the "AI" feature, which makes copying and pasting much worse and annoying. We need an option to disable this at least on a per-user base. Right now this feature is actually making Jira worse to use for us.

We are very similar to 3b21f807ea4a. We have certain users with access to the tools that should not utilize AI since they need to be limited to only the details within the work item itself. If they need to write new content, it needs to use one of our own pre-defined templates. And other users that are working with sensitive data that should not be read by anything or anyone else. Without segmentation by users/groups, unfortunately no one can use Atlassian Intelligence at all across our entire Org and it needs to remain fully off across multiple instances of Jira/Confluence.

Separately, we only need certain features to be on for most of the user groups, so it would be helpful to have user/group permissions for each AI feature. But to start, as long as we can turn all features off for certain users and groups, we should be able to unblock AI for many thousands of our users.

As a response to the updated request description:

• Name of the organization, your role
  • I am one of our full time Atlassian admins, and have the overall responsibility for our Atlassian Cloud products. I am the "Atlassian product owner".
• Why do you need user/group level control? What impact will you have if these controls aren't provided?
  • We need user/group level control because for some content in our Confluence site we are only Data Processor, and not Data Owner, and some of that data is not allowed to be handled by AI or processed outside of our site/region, or by non-disclosed third party data processors. Having these restrictions available to us would allow us to restrict AI use for those user that have access to sensitive data.

Or in other words "Because of GDPR".

• How are you using AI today in the absence of these controls?
  • We are simply not. We cannot enable it because we are lacking better administrative control and tools.

As a bonus note.
For us AI-110 would resolve our problems in a better way, but since this one for some reason has a higher priority I am making myself heard here.

Being able to restrict access to AI features for groups of users would be good.
That would allow us to at least restrict AI access for people who have access to sensitive data that we can't have AI process.

Speaking as a user, I find all the underlined words an annoying distraction, not helped by the fact it keeps highlighting internal part numbers with wrong information.

I would like to turn it off for myself, with no admin privileges and without having to raise a support ticket or bother our admins.

I should be able to click on one of them and have a 'never show me this again' option.

[I am speaking for myself, and my opinions and preferences do not represent those of my employer].

If it's enabled (generally), users should be able to disable it for themselves as well. This was done for Confluence, so it should be possible for Jira.

Some users find the AI features useful and like them. Other users think the AI does not help and just distracting in the UI.
Can we please have a way to disable them per-user?