I agree with your point and have long disliked when Firefox squanders their limited resources on side quests - which is too often. But, to 'steelman' their motivation to 'do something' on AI, this analysis article sums up why major AI and browser vendors are pushing : https://entropytown.com/articles/2025-10-31-openai-atlas-ai-....
I just think Firefox is taking the wrong approach. Trying to run with the pack of large commercial entities supporting their multi-prong corporate agendas does nothing for Firefox long-term (while annoying their users and looking like a buzzword-chasing 'me too'). This is a perfect example of when Firefox should zig instead of zag. Per the article I linked:
> “the hard part of an AI browser is not chat, it’s process and trust isolation.”
Instead of feature parity on AI, Firefox should race to technically position with APIs as the friendliest 'host browser' for AI companies outside the big five (eg "everyone else"). That gets some AI vendors actually recommending Firefox as the "works best with..." option instead of ignoring FF. Plus AI projects, researchers and LocalLLama-type hobbyists will be attracted. Sure, that's currently a small segment but they have high-potential for growth. It's very early days and today's AI leaders may not be tomorrow's AI giants.
I just think Firefox is taking the wrong approach. Trying to run with the pack of large commercial entities supporting their multi-prong corporate agendas does nothing for Firefox long-term (while annoying their users and looking like a buzzword-chasing 'me too'). This is a perfect example of when Firefox should zig instead of zag. Per the article I linked:
> “the hard part of an AI browser is not chat, it’s process and trust isolation.”
Instead of feature parity on AI, Firefox should race to technically position with APIs as the friendliest 'host browser' for AI companies outside the big five (eg "everyone else"). That gets some AI vendors actually recommending Firefox as the "works best with..." option instead of ignoring FF. Plus AI projects, researchers and LocalLLama-type hobbyists will be attracted. Sure, that's currently a small segment but they have high-potential for growth. It's very early days and today's AI leaders may not be tomorrow's AI giants.