For context, I have been using a pretty niche OS on my phone for a long time but recently the community and the 3rd party appstores are flooded with LLM content - and fans. Its ethical, geopolitical and societal dangers are handwaved away at best, or ridiculed or countered with stupid non-arguments ("you wouldn't tell a farmer not to use a tractor").

Everybody is in a hurry to churn out the next pretty port of an Android app.

The actual OS devs have not taken any sort of public stance, not even in comments.

And it got me thinking. If this continues maybe LineageOS & F-Droid (no Gapps of course) might turn out to be a better alternative again, despite it being essentially Android and Not True Linux.

But how do they position themselves with AI?

If you want to suggest a 3rd option, please do so - but only if it works reliably as a daily driver - incl. telephony and messaging - on at least a small number of consumer devices that are no older than 8 years.
And only if they clearly position themselves against AI in code and generally.

Not sure why you are asking this in a linux phone community but this is the generative AI policy of flathub, pretty much the store for linux phones:

Generative AI policy

Submitters must disclose any AI-generated code, documentation, packaging, or other material they know or reasonably believe is included in the application or its Flathub packaging. The disclosure must identify the affected parts and approximate extent.

Flathub manifests must not contain AI-generated or AI-assisted content. Disclosure does not exempt manifests from this restriction.

AI used only for research, discussion, or debugging does not need disclosure when no generated material is included in the application or its Flathub packaging. Other disclosed AI-generated material is evaluated at reviewer discretion. Reviewers may reject a submission, including without further review, based on the extent or role of generated material or concerns about its review, quality, or maintainability. Disclosure does not create a presumption of acceptance.

AI tools or agents must not open or automate Flathub submission pull requests, or generate their commit messages, descriptions, review comments, or replies. Submitters must not request AI-agent reviews.

Undisclosed or materially misrepresented AI-generated material, or prohibited AI-generated submission or review interactions, may result in rejection. Repeated violations may result in a permanent ban from future submissions and activities.

https://docs.flathub.org/docs/for-app-authors/requirements#generative-ai-policy

In general also outside AI specifically the requirements are pretty strict.

The only one that caught my attention recently was GrapheneOS being in favor of using LLMs with the argument of "Linux kernel and other projects are using them already anyway". Which is a strange argument for an OS that positions itself (and is known as) as a better alternative to Android.

Et tu, Brute?

It is strange that their paranoia allows them to use LLMs at all.

The AI hype is a sickness, and it's spreading.

You'll get better answers on an android community rather than a Linux phone community

Answering myself wrt F-Droid:

tl;dr: the situation appears very similar to what I describe in my OP.

• They seem very focused on fighting Google's recent shenanigans and getting funded, and FOSS of course.
• I found no mention of AI whatsoever in their CoC
• I found no mention of AI whatsoever in their list of anti-features, and this seems to be the biggest indicator to me.
• Not in their Blog either (by title, going back one year) but my search engine gave me this article, which seems to have been removed since: Home much of F-Droid is LLM-generated.
• There's a forum post asking a similar question with some very predictable drama. Straight out the door some "contributor" ridicules OP's reasonable question in that typical, passive-aggressive smiley emoji coder way. Just what I have experienced. Thankfully other "contributors" at least see the value in pondering these things. While the question is never fully answered, the thread contains some interesting links to how other FOSS projects position themselves.

I found an important clue in the latter:

the watered down "Open Source AI Definition” that Mufelli pushed through the OSI.

https://opensource.org/ai/open-source-ai-definition

Straight away I don't agree with it:

Why we need Open Source Artificial Intelligence

I'm guessing this is to be seen as a quasi-official stance to AI that most F(L)OSS will adopt or at least base their own policies on...

I did notice that, seen strictly logically, there's an error in my posting this here, but I still feel it belongs.

Anyhow I did do that already