> Unlike a purely generative AI model, which is prone to spout unsubstantiated assertions, we trained our large language models (LLMs) to process multiple sources of information present on the Web ... In addition, the provenance of original sources of data is cited at all times via links. This maintains the rightful attribution of information, and helps users assess the trustworthiness of the sources, both of which are needed to mitigate the authority biases of large language models.
At least it has sources and citations rather than outright generating bullshit out of no-where and is only used for summarization. Good.
It would be much worse of a solution if there were no sources from the output. A better LLM-based search engine who certainly need to cite their sources and explain why they chose them to reduce bias, otherwise it is untrustworthy to use in general.
> Although the industry is generating much hype around AI, at Brave we are not yet convinced that LLMs can replace search as we know it. However, if used properly, these new models can help the user navigate results, which is the approach we follow with the Summarizer.
Summarization of existing facts is the only safe use case for LLMs that makes perfect sense, as long as the sources are disclosed which shows a glint of transparency, unlike the hype from Bing and OpenAI have done nothing but cause wanton chaos and recklessness with their creation(s).
Brave seems to be taking the best approach in using LLMs with responsible AI and doing that more seriously than OpenAI and Microsoft is.
They also provide a little detail about the models used:
> The base LLM models are based on either BART or DeBERTa (which are open source and hosted on Hugging Face), with heavy retraining based on our own data from search results
At least it has sources and citations rather than outright generating bullshit out of no-where and is only used for summarization. Good.
It would be much worse of a solution if there were no sources from the output. A better LLM-based search engine who certainly need to cite their sources and explain why they chose them to reduce bias, otherwise it is untrustworthy to use in general.
> Although the industry is generating much hype around AI, at Brave we are not yet convinced that LLMs can replace search as we know it. However, if used properly, these new models can help the user navigate results, which is the approach we follow with the Summarizer.
Summarization of existing facts is the only safe use case for LLMs that makes perfect sense, as long as the sources are disclosed which shows a glint of transparency, unlike the hype from Bing and OpenAI have done nothing but cause wanton chaos and recklessness with their creation(s).
Brave seems to be taking the best approach in using LLMs with responsible AI and doing that more seriously than OpenAI and Microsoft is.