Bing tests AI Deep Search that turns your simple question into a detailed prompt
Microsoft is readying a new Bing feature that should take the hassle out of coming up with your own AI prompt. The GPT-4-powered capability, called Deep Search, takes your Bing query and expands on it, allowing the search engine to find answers about several topics related to your question on the web.
As an example, Microsoft shows how Bing turns a vague search for “how do points systems work in Japan” into a detailed prompt that asks Bing to:
Provide an explanation of how various loyalty card programs work in Japan, including the benefits, requirements, and limitations of each. Include examples of popular loyalty cards from different categories, such as convenience stores, supermarkets, and restaurants. Show a comparison of the advantages and disadvantages of using loyalty cards versus other payment methods in Japan, including current rewards and benefits. Highlight the most popular services and participating merchants.
Bing may also expand that search to include descriptions of immigration policies and public transportation fares, too, as GPT-4 attempts to “find all the possible intents” of a query.
Whatever topics Bing comes up with will be displayed in a panel on the search results page. You can then choose which one best fits what you’re looking for, and Deep Search will curate its search results by ranking them “according to how well they match the comprehensive description.” Microsoft says conducting a Deep Search could take up to 30 seconds, but if you don’t have that much patience, you can always just continue using Bing’s standard search feature.
Microsoft revealed Deep Search as part of its announcement about Copilot’s support for OpenAI’s GPT-Turbo model. While Deep Search is currently an experiment that’s only available to a small group of users, I’m looking forward to trying it out. Sometimes coming up with a prompt for AI is a task within itself, which kind of defeats the purpose of using it for productivity.
Read the full article Here