Article Summary:
Describe new AI enhancements in Firefox 152.0 and their state in Fedora Firefox package. What is disabled / enabled by default and how to change the defaults.
Article Description:
I have read and understand the Ai-Assisted Contributions Policy
For Editor Use Only
Editor:
Image Editor:
Publication Date: June 16
Preview Link:
This sounds like a good news article that many Fedora Magazine readers would be interested in. +1 from me.
Upload your article to our WP Engine instance and, when you have it ready for review and publication, let us know with a follow-up comment here.
Thanks for contributing to Fedora Magazine!
I found this page yesterday by accident on a search engine. As it’s indexed, I assume it is ok to comment (I write as user, not as moderator).
I hope I am allowed to add some thoughts, about what might be interested about firefox/ai for users (there is much discussion I find about it, incl. articles, but nothing definitive, a lot of speculation, and in the end, I can only say that Firefox is currently quite intransparent).
I accidentally found out of the “AI measures” and "AI control"on Firefox (didn’t invest time in getting into the debate in the past), and became skeptics when reviewing the legally precise formulations that use trigger words like “block ai” but effectively re-interpret the term to just “hide” rather than “disable” any ai.
Following this, I end up at legal texts and FAQ that effectively just repeat this. Then I found the solution for advanced users: about:config has the solution. As most users do, they just search all the ai-settings (I think about 10), disable them, and then “fire and forget”. However, others seem to found already out that Firefox automatically re-enables all of these ai-settings at updates. While I did not verify, the mods of Mozilla seem to have not questioned this, but confirmed that there is no general way to disable it permanently (maybe that is already obsoleted? Or was a bug?). That gets finally complemented by a change of terms in 2025 that seem questioned by many media and lawyers about firefox taking ownership of data produced of/with me, but they are selective and not neutral either, and the background of the new terms remains effectively unknown.
At the moment, it is mostly confusing intransparency I wonder about, but the combination of the legally-precise formulations of “block=hide-but-not-disable” with the auto-reset of auto-config make me feel a little like Firefox tries to mislead me in hostile ways and wants to produce some data of/with me but keeping me unaware unless I support it. However, what data that might be, no clue, as it doesn’t really make sense in context.
That said, ignoring my first subjective feeling, I can also think of several rational and “friendly” explanations that can explain the three issues in context.
A side note from my mod experience:
A debate is there. It is not yet in Fedora, but given how people currently respond to AI and privacy topics, I fear the next topic that is screaming for “moderate me hourly” when it is started by irrational interpretations of the reality, filling the gaps as hostile as possible and creating a corresponding discussion.
I would be happy if the debate about firefox, enforced ai (may that be true or not?) and privacy, is started by you, with rational facts, your background information, and setting the issues in context (in general and with each other). If you start a debate, you have the power to shape its beginning in a rational way. If someone else does it later based on whatever accidental incentives, the outcome is unpredictable.
Starting the debate means you have the power to shape the initial social dynamics of people (shape the “group psychology”), while others later can only try to influence it. That can be used to create a useful debate (incorporating already important data available to you) from the beginning, rather than reddit speculations as point of starting.
For me, it would be interesting to have a clear and conclusive (and reproducible) explanation of the three current issues (“block ai”=hide-but-not-disable definitions, legal terms, about:config auto-re-enabling), but in context to and with each other (maybe there is even more?).
I lack time to get into this, and would prefer if Mozilla/Firefox people put their facts in context and give me a conclusive and reproducible explanation, rather than me being forced to speculate in a sea of possibilities.
I love the QA integration and the strong feedback loops we have with Firefox, and the strong review/testing it has, which is why I keep using it I think for 20 years or so. But I admit, the overall outcome of what I observed yesterday, without conclusive reproducible explanation, I couldn’t avoid some thoughts like “do we have alternative browsers, just in case?”.
Talking as subjective user again: if our Firefox team would also start to engage in promoting these ai means but dismiss/ignore the intransparency and the related questions about the mentioned developments, I think it would rather increase my skepticism.
Just some thoughts, and a perception that others might or might not share. Don’t know how many that would be interesting to. But at least I would be happy to read that 
In any case, thanks for all your contributions over time, and looking forward to an interesting article, may it be with or without the points I mentioned!