hi team, for the same KPI inside a Liveboard in different orgs, some users see Ask-AI but some users see Explore in the tiles. Is there a reason for it? They both have same priviliges. https://cdn.discordapp.com/attachments/1143209464875462756/1532592071389610075/image.png?ex=6a6d6918&is=6a6c1798&hm=2ebfa82278ad23844a6d11b14335113d9c0a66bc99c9fe8b408ffdd5db3f1c22& https://cdn.discordapp.com/attachments/1143209464875462756/1532592071678890044/image.png?ex=6a6d6918&is=6a6c1798&hm=a3f996de631faf206432a69db6622be4d86ff96ef217658d27152945638acb49&

message has been deleted

Yes — this is usually feature enablement / org-level rollout, not tile-level privileges. The official docs show that Ask AI on Liveboard visualizations was introduced as a feature that can be turned off by default and enabled separately. The docs explicitly say the Ask Sage / Ask AI capability must be enabled on the instance, and for embedded users it can also be controlled with an enablement attribute. So if the same KPI tile shows: - Explore for users in one org - Ask AI for users in another org the most likely explanation is that AI/Ask AI is enabled for one org/tenant context and not the other, even if the users have the same object permissions. Same privileges on the Liveboard do not guarantee the same tile action if the underlying org, release/feature flag, or AI capability enablement differs. What to check: 1. Whether Ask AI / Ask Sage / Spotter-in-Liveboard is enabled in both orgs. 2. Whether both orgs are on the same cluster/release configuration. 3. Whether any embed setting or org-specific config is hiding/showing the action. 4. Whether Support enabled the feature only for certain orgs or tenants. The docs state this capability was not simply universally on by default. I couldn’t find a doc statement saying “same privileges should always produce the same Explore vs Ask AI action.” The documentation points more strongly to feature availability/configuration differences than to user privilege differences. You are interacting with an AI system. Responses should be reviewed.

@spotter-code where can i check if its enabled or not?

Check it in Admin / feature settings for the instance or org, not just in the user role. What the docs confirm: - The Ask Sage / Ask AI capability is a feature that is off by default and may need to be enabled by ThoughtSpot Support on the instance. - For embedded users, it can also be controlled with the

enableAskSage

attribute. - If it is not enabled, users will typically just see the normal Explore action instead. So I’d verify these 3 things: 1. Org-level / instance-level feature enablement - Ask your cluster admin / org admin whether the AI feature is enabled in both orgs. - If needed, confirm with ThoughtSpot Support, because the docs explicitly say this feature may need Support enablement. 2. Embed config - If this is an embedded Liveboard, check whether the app is passing:

js
   enableAskSage: true

If one org’s embed path sets it and another doesn’t, that would explain the difference. 3. Same cluster / same release - If the orgs are on different environments or rollout states, feature availability can differ even when user privileges match. I couldn't find anything in the official ThoughtSpot documentation for a specific UI page that shows “Ask AI enabled/disabled” status per org. Closest match:

enableAskSage

— embed setting used to enable or disable Ask Sage for embedded application users. You are interacting with an AI system. Responses should be reviewed.

@spotter-code its same org and same configs in the embed

If it’s the same org and the same embed config, then it’s probably not privileges and not a simple embed flag. Most likely causes: - **Feature rollout / AI entitlement at user level**: the docs confirm

AskAi

is a distinct action used for Spotter/AI experiences in Liveboards, but they do not document a per-user privilege that directly maps to “show Ask AI vs show Explore” in that tile CTA. - **Different user experience state or backend eligibility**: if one user/session is eligible for the AI flow, the tile can show **Ask AI**; otherwise it falls back to Explore. I couldn’t find official docs that list the exact decision tree for that button swap. So the practical places to check are: 1. Whether Ask AI / Spotter is enabled on the instance 2. Whether those users are licensed/entitled for the AI experience 3. Whether the same worksheet/KPI is AI-eligible for both users 4. Whether a phased rollout / tenant-side feature flag differs by user session I couldn't find anything in the official ThoughtSpot documentation for the exact rule that decides why one Liveboard tile shows Ask AI and another shows Explore for users in the same org with the same embed config. Closest match:

Action.AskAi

— action ID for launching the Ask AI / Spotter experience in Liveboard and embed contexts. You are interacting with an AI system. Responses should be reviewed.

@spotter-code https://developers.thoughtspot.com/docs/Enumeration_Action Whats the difference between the https://developers.thoughtspot.com/docs/actions

They’re different docs for two different things: -

Enumeration_Action

= the authoritative SDK enum definition. It lists the actual

Action

members you use in code, such as

Action.Explore

and

Action.AskAi

, plus their exact string values and descriptions. -

/docs/actions

= the usage/catalog page. It explains where actions appear in the product UI and which embed types support them, grouped by page like Liveboard, Search, Answers, etc. So in practice: - Use

Enumeration_Action

when writing code with

visibleActions

,

hiddenActions

, or

disabledActions

. - Use

actions

when you want to know whether an action applies to a Liveboard, Answer, Search page, full app, etc. One important detail from the enum page: -

Action.Explore

= “The Explore action on Liveboard visualizations” -

Action.AskAi

= “Action ID for the Ask Sage in Natural Language Search embed, Spotter in Liveboard, full app, and Spotter embed.” That means

Explore

and

AskAi

are distinct built-in actions, not just two labels for the same control. You are interacting with an AI system. Responses should be reviewed.