bleepingcomputer[.]com/news/security/brave-search-launches-ai-powered-summarizer-in-search-results/
I've stayed way from Bing's AI search, since *to me* the value of a Google search is the ability to review what several sites have to say, and then go with the consensus -- too often a web site is just wrong, and that's ignoring all the cases where someone's politics &/or biases rear their ugly head. And *to me* there's just not much value in using AI search when you have to double check the results to see if they're accurate or not, and that's assuming the AI itself doesn't make an error, which is currently a possibility. Sure it can mimic a personality, & lots of folks have had fun [and some success] trying to drive it nuts, but there's a difference between work & play.
According to Brave:
“Unlike AI chat tools which can provide fabricated responses, the Summarizer generates a plain-written summary at the top of the search results page, aggregating the latest sources on the Web and providing source attribution for transparency and accountability”
In BleepingComputer's tests, the Summarizer does not appear at the top of search results when performing simple queries but rather when searching for the answer to a particular question.Apart from the summary at the top of the search results, the tool will also replace the standard result excerpts with a summarized version that will theoretically better highlight the answer to the user’s query.
Brave uses 3 AI models that limit themselves to [hopefully] being useful:
Question answering – the model tries to extract answers from text snippets.Results classification – categorize result candidates based on various criteria such as hate-speech, vulgar writing, and spam.
Result processing – Rewrite the final set of the candidate text to remove repetition and improve readability.