Hacker News Reader

Comment by breather | Hacker News Reader

"AI"

Well, the company is fucked. I wonder who will pick up the pieces of the browser tech.

There are two pieces of tech that AI might apply to:

1) Semantic ad-blocking (this will fuck up their only source of reliable revenue so it won't happen)

2) content filtering for a child-friendly mode. This actually is a good idea, but it won't happen because it'll be easily repurposed to perform bullet point #1.

View on HN

Firefox is already using AI (fully local) for their new translation feature.

A) I don't think that's the type of feature to which they are referring to, B) I wouldn't call that "AI" by the current sense of the word when it can be narrowly described with "automated translation", and C) we've had high-quality translation for well over a decade now with or without Firefox.

Firefox's translation is based on LLMs not far off from the AI chatbots in the core tech. It's also nicely local to your machine.

I wasn't aware of that, thats actually pretty cool!

Gotta one of the first legitimate uses for AI outside of generating semi-passable scripts and email templates.

That's all very nice, but it doesn't change anything i said.

#2 is just a document classification task, you can do it with DistilBERT.

> #2 is just a document classification task, you can do it with DistilBERT.

I'm assuming there's still an immense amount of effort into ensuring quality and then packaging and distributing it, especially if you're selling as for children, especially if you want it to work on a phone without sending the browsing data to a third party.

Yup, and that’s the thing with the LLM craze - it’s super energy intensive and for the most difficult problems you still need really good hand curated data anyway.

My point is more “we know how to do this already and it’s just a hard problem”.

None of the above. Mozilla is currently working on... chatbots.

Reply on news.ycombinator.com