There is a open issue where they want to investigate the ability to have a opt-in to contribute interactions for fine tuning and a point is raised there that the key thing to consider is transparency and and what consequences end users will have when choosing opt in vs opt out.
And like rewards or contributing to web discovery in search this feature will obviously (i´m assuming) be opt in, meaning if you don’t want to use it you can leave it disabled.
Personally on one hand i´m excited to see them do something new but i really have not kept up with this whole AI/ChatGPT/LLM more than not using any due to privacy concerns and that i really have not been able to find a reason to use one so currently i don’t really have any opinions on this either way.
I would appreciate to hear some of your thoughts, concerns, reservations etc?
And please let´s keep the comments on topic and refrain from cluttering this up with posts like “this is why i use browser X instead”.
In regards to Brave (as a company), I think this is a good move. Like it or not, AI is growing, and it’s being integrated into products/services people use. To compete, they need this functionality.
Would I prefer that I use a self-hosted option? Absolutely. Will I spend time and electricity training one? No. If this is ethically collecting data, and is (somewhat) privacy respecting, it could be a great addition. That being said, I’d like to see it in action before I give praise or negativity.
Leo, Brave’s browser-native AI assistant, is now available in Nightly version for testing Source
With regards to privacy the article states:
A note on anonymity
Chats in Leo cannot be used for training purposes, and no one can review those conversations, as they’re not persisted on Brave’s servers—conversations are discarded immediately after the reply is generated.
And comparing that to the Leo wiki page where the same section states (I have marked in bold the parts that are missing from the article):
Leo is free to use for any desktop Nightly user, and no user login or account is required. Chats in Leo cannot be used for training purposes, and no one can review those conversations, as they’re not persisted on Brave’s servers—conversations are discarded immediately after the reply is generated. For this reason, there’s no way to review past conversations or delete that data—it isn’t stored in the first place.
the next part is identical on both the article and wiki:
What data does the Brave browser send?
If you use Leo, the browser shares with the server your latest query, your ongoing conversation history and, when the use case calls for it, only the necessary context from the page you’re actively viewing (e.g. the article’s text, or the YouTube video transcript).
I do wonder if Leo is planned to be a Premium feature?
If not it could be a opportunity for Brave to add Leo as one. So Summarizer would continue to be free but to get the actual conversational thing going it would require Brave Search Premium which seems to currently cost 3 USD/month.