The other devs were generally against any form of disclosure, saying that they didn’t want to ‘shame’ developers who used it.

The discussion was locked and then removed from public view due to public backlash.

Solus mostly has just a skeleton crew left after the creater, Ikey Doherty, abandoned it suddenly. I personally didn’t think it’d be around long-term even before this new policy.

The most effective supplement with the biggest impact I’ve ever tried is plain old magnesium oxide (the expensive forms are a waste of money, and you’ll actually absorb less elemental magnesium from them. Just take the oxide form with a meal to avoid the shits).

It’s difficult to get enough magnesium in most diets, and I found that ensuring adequate magnesium intake had the effect of making my mind just feel sharper (that effect is especially prominent if you’re supplementing vitamin-D as well, since that tends to use up magnesium in its conversion process, IIRC).

nothing else has had such a noticeable, non-placebo benefit (except Vitamin D if you’re deficient, which will make you crazy depressed without getting sun or supplementing).

Personally I thought that was a nonsensical argument from them, as they could’ve just put in the policy to declare that AI was used, but just not mention the specific one. Better yet, IMO, would’ve been to ban all non-self hosted AI use, which basically cuts out all the corporate AI’s.

Though even with the self hosted models available, the ethical and copyright risk remains, so I’d still have personally opted for a total ban.

Movim might be a good option.

Deltachat has an extremely smooth on boarding process and is the same app for all platforms, but it cannot do voice calls or screen share (you can leave voice messages though).

Jabber/XMPP generally has a less smooth UX, and different apps across different platforms (except the Movim client, which is a web app that works on all platforms, but doesn’t do reliable notifications on mobile).

With the Movim client, Jabber/XMPP is basically a complete Discord and Signal replacement. It can do group video calls, screensharing with audio (only with chromium based browser for now), as well as voice and text channels under different servers.

Regarding metadata, you may want to read the FAQ.

It will have forward secrecy after they finish implementing Autocrypt2.

A person also has to determine their security threat level. If they think a state actor may take a personal interest in them, then they’re probably best using in-person offline communication. If they must communicate online, then IMHO I don’t think Signal is a good choice due to it being centralized (everything goes through their servers, and they are a single entity making them easier to infiltrate or coerce by governments) and requiring you to link a personally identifying number or credit card.

If they just don’t want a corporation datamining them to sell to advertisers, Deltachat is more than good enough.

For the writers out there, NovelWriter is a fantastic FLOSS alternative to Scrivener (has similar organizing features), and it entirely uses markdown.

The VA part was especially noticeable in CP2077, the male VA sounded like someone doing a stiff stereotypical mobster accent, while the woman VA was excellent, with far more emotion put into the lines, couldn’t imagine playing it any other way.

False. There are a number of models trained on data they were explicitly given permission to train on.

So far from what I’ve seen when I investigated self-labeled ethical models, I found that they weren’t actually trained ethically, and not on code that is compatible with GPL or code that doesn’t need to be attributed. (if you can find one that has proven otherwise, I’d be open to taking a look at it).

That said, I don’t think there would have been as much push back if the KDE proposal had explicitly limited AI use to self-hosted models.

There’s no reasonable way they can actually ban this. And they don’t have the time or money to police it.

According to the KDE devs, the proposal was intended to allow them to easily reject clearly slop PRs. A policy that limits AI use to self-hosted models would have achieved that goal just as well as one that allows any model, and would essentially be an honor system, which I suspect most contributors would be willing to follow (obviously not all, and there is nothing stopping a contributor from lying that their corporate AI code was generated on a local LLM, but then again, there are also rules asking contributors to not knowingly submit copyrighted code from proprietary software, which is also an honor system).

You are simply misattributing them to artificial intelligence and not wealthy authoritarians seeking to profit. By all means, fuck zuck, musk, altman, etc and their companies. They’ve always been and always will be destructive in every way possible. AI or not.

To be honest, I’m struggling to understand why you would be against a policy that at least attempts to limit the benefit/gain to those wealthy authoritarians. A policy that did so would not require any further resources from the dev team compared to the one they attempted to implement, it would basically just be an official request to contributors to not use the explicitly fascist owned tools that are most contributing to global warming and our subjugation.

The biggest contributor to the backlash of the proposed policy was the additional aspect explicitly not requiring AI used in a PR to be disclosed, regardless of its quality (and which is unrelated to the ability to quickly reject obvious slop PRs).

AI can be implemented ethically.

Potentially, yes. Currently this does not exist.

AI can be done without destruction of the environment

The AI policy proposed did not restrict AI usage to non-destructive self-hosted models.

Who cares if AI replaces people as long as we take care of the people.

The people who are most impacted by the negative externalities of AI are not taken care of. The people losing their jobs are also not taken care of. That is why people care so much. It’s not viable to handwave away the negatives AI are causing in the here and now with the assurance that in some far off day we’ll live in a post-capitalist world where the negatives won’t matter.

Yes OP has been daily Doom posting for the last month or so.

I’ve posted about the AI policy of KDE and two follow-ups as the situation developed over the past 4 days, and am unlikely to post anything further on KDE until a new development occurs, which I anticipate will be a while.

I have previously posted here once about Xfce using AI in their Wayland compositor, and about Debian adopting a permissive AI policy when the outcome of that vote occured.

Deltachat and Jabber, both decentralized encrypted messaging apps (as opposed to Signal, which is centralized), don’t even require an email to create an account, let alone a phone number or payment.

A good federated alternative to Discord is Movim.

People suspected he’d uploaded Fluxer in one big chunk to hide LLM usage. He admitted to using some AI help initially but claimed that he would use it sparingly.

This new issue is that he seems to be using LLMs heavily, based on the fact that he uploaded a million lines of code last month alone.

https://github.com/fluxerapp/fluxer/graphs/contributors?from=8%2F22%2F2026

If you didn’t get a chance to see the original policy proposal thread that started the controversy, you can still read it here.

There are serious concerns beyond pure code quality.

Xfce and Debian have long track records as well, yet both have adopted permissive AI policies.

The way Nate is framing the original intent of the proposal, that KDE was swamped with AI slop and they just needed a quick policy that would allow them to comfortably reject it, doesn’t seem quite accurate. If that was all they needed, it seems like it would’ve been quite easily to propose an ad-hoc policy of allowing maintainers to reject suspected AI slop PRs at will, which would solve the immediate problem while allowing them to gather more feedback on a ‘proper’ policy. Adding a provision to not require any AI disclosure doesn’t help with the stated goal of just wanting a way to easily reject slop.

I’ve read the positions of some of the developers involved in the proposal thread on other platforms, and most have made it very clear that they are pro-AI (I do not include Nate among them, as he has stated he actually dislikes AI and doesn’t use it). Among that group, the concerns about AI ethics and their use are, ultimately, of secondary concern over the prospect of increased productivity and time savings. That inclination is what resulted in so little credence given to the people raising concerns against it in the proposal thread (it was clear that the concept of not accepting AI code, or even just limiting it to self-hosted models, was not on the table from the outset).

Nate is clearly distressed from the negative reaction the proposal has caused, and I can sympathize with that. However, from what I’ve read of his responses on his linked blog, IMHO it is mostly damage control, and still isn’t addressing the main concerns people had about the original policy proposal. When people point out the copyright, environmental, and ethical issues, Nate will acknowledge those, but insists that the original proposal, by virtue of it allowing them to quickly reject obvious AI code, would’ve been an improvement (despite it also giving the provision that AI use didn’t have to be disclosed, essentially meaning that those skillful enough to make it not obvious will still be perpetuating the ethical, environmental, and copyright issues, but just in the dark without anyone knowing the severity).

In the linked blog post, many of the comments are saying that KDE effectively closed down most avenues to provide feedback on the proposal, thus leaving them with essentially the only option of making the KDE For People open letter, as otherwise it appeared that any public dissent against a permissive AI policy was ultimately going to be dismissed as the Pro-AI group decided on a policy behind closed doors.

Their justification for planning to make it a closed door decision was two people in the proposal thread, both of which easily could’ve been removed from the thread, or the thread closed to developers only, and their discussion could’ve continued without any difficulty. I personally don’t believe their justification, as it clearly seems like it was closed and ultimately made hidden to avoid the public relations fallout the main discussion was having for being so dismissal of the AI concerns brought up.

I invite anyone curious to read the blog and the comments below, and come to their own conclusion. (Be sure to click ‘old comments’ on that blog, as most are now hidden behind that).