Microsoft has one AI-powered update to its Bing search engine. It’s not surprising, since we’ve lived with the same algorithmic approach to search since Google launched over 20 years ago, so any innovation will turn heads.
Microsoft CEO Satya Nadella called it “a Mosaic moment” – referring to the early Internet browser and pointing to a very interesting future for software. Before the Mosaic browser, the World Wide Web was simply a way to explore and share documents; after the launch of the early browser in 1993, it became the user interface for the wider internet. AI in Bing is much the same, as it is the start of a series of experiments that could change the way we interact with software.
How Microsoft delivers AI-enhanced search
At the heart of Microsoft’s AI-enhanced search is the same co-pilot assistive AI concept it uses with its developer tools. The AI isn’t what drives the process – the user does. Switch to the chat feature and you get a way to fine-tune your questions and manage how they appear.
For example, I used it to generate a set of instructions for a relatively complex function that I’ve struggled to use on a new camera. Not only did it give me a clear explanation, but I was able to quickly convert that result into a step-by-step guide and then into an email. Results are supported with references, so you can go directly to primary sources.
Possible use cases for this AI-enhanced search
There’s something important here for organizations looking to use tooling like this for customer support. Users can ask the questions that matter to them and get data the way they want it. Instead of an FAQ, the interface to your manuals and support database becomes a browser sidebar or a search query.
It’s the same for shopping and ecommerce, where the search tools build comparison and recommendation tables in response to user prompts, returning results close to Bing’s original mission as a tool to help you make decisions.
Microsoft’s secret sauce: Prometheus
Key to Microsoft’s approach is what it calls its “Prometheus model.” Named after the mythological hero who stole fire from the Greek gods, it aims to keep people informed, building on Microsoft’s work in responsible AI. A bad early experience with a chatbot Microsoft has learned many lessons, and its work to ensure AI is secure is ingrained in the company from the board level.
Prometheus is more than a new version of existing tools such as ChatGPT: it combines OpenAI’s large language models with Microsoft’s experience with Bing and with its responsible AI framework. While OpenAI’s demo services only take input from a text box, Microsoft adds additional context, including location, date, and time, while splitting searches into multiple parallel queries, with the goal of speeding up requests and improving the quality of results. Results are parsed and used as inputs to the model, looking for new insights that can drive additional searches, an approach Microsoft described as a “virtuous loop.”
The first implementation of this AI tooling is in both the Edge browser and the Bing search engine. Microsoft provided a preview of how the service will eventually appear in the rest of the company’s productivity tooling. In Edge, a new Bing sidebar lets you perform searches outside of the traditional browser, giving an idea of how it could be implemented in the familiar Office applications. The new sidebar design that Microsoft has rolled out in Office is very similar to the Bing AI tooling being trialled in the developer version of the Edge browser.
While Microsoft hasn’t yet announced any plans to provide API-level access to this tooling, it would be surprising if it wasn’t on the company’s radar. Microsoft is first and foremost a platform company, and its use of generative AI clearly indicates what it thinks its next platform should be: assistive AI.
TO SEE: Artificial intelligence ethics policy (Tech Republic Premium)
An AI platform for businesses
What would a supporting AI platform for Microsoft 365 and Dynamics look like? More importantly, how would it work? Microsoft’s Prometheus tooling is designed to provide a series of guardrails around AI tooling, with a focus on what the company called “prompt engineering.”
If you look at early platform tools like the recently launched Azure OpenAI, the API for a generative AI like GPT is very simple: it’s a way to send a string of text to the service. What’s important is getting that string – the prompt – correct.
It’s also interesting to see how the platform helps you find that prompt. With Bing’s OpenAI tools, it’s better to start with a relatively open question and then refine your results. It’s an interactive platform designed to help you find what you’re looking for, and the best results come from those interactions. Simple searches are handed over to the traditional search algorithms as there is no point in using the computing resources required for an AI powered search.
Then there’s the bigger issue of training data. While large language models like GPT are trained on public data, that’s less important when it comes to working with the data stored in your own applications. Again, Azure OpenAI provides some guidance, with the option to use a custom model that combines the existing GPT LLM with a selection of your own data. Both Dynamics and Microsoft 365 have their own data models, which would work well as part of a comprehensive Prometheus model, especially since Bing already uses that data in company search mode.
A contextual computing future
Near the end of his tenure as CEO of Microsoft, Bill Gates talked about the idea of contextual computing, where systems responded to the context around a user. It’s not too hard to describe Bing’s AI copilot as a contextual computing tool, using it to refine searches and results without leaving the context of the original query. Tools need context to work, and while that context is limited to the scope of a query for now, it’s clear there’s scope to go much further.
Asking a question about a place might give you a traditional trip output, but narrowing it down might give you an itinerary for where to stay and a list of places to eat each day. Links in the answers let you open new pages, while references in Bing’s new chat interface show where those results came from. No context is lost throughout the series of questions – you focus on the AI’s answers to get the results you want. The service can also reformat data and produce tables to help compare the results.
It all comes down to a very different way of working with both search and the web, prioritizing the information users are looking for and presenting it the way they want it. Like the first graphical web browsers, it’s still very new and not fully formed yet, but it’s clearly not something we can ignore. This technology will change the way we build and structure content in the future.
TO SEE: Hiring Package: Architect for Artificial Intelligence (Tech Republic Premium)