From new copilots and AI improvement instruments, to vector search and AI chips, synthetic intelligence featured prominently in Microsoft’s annual Ignite builders convention held this week. It additionally unveiled some knowledge information round OneLake and Microsoft Cloth.
It will be an understatement to say that Microsoft is bullish on copilots. “Microsoft is the Copilot firm,” the corporate claims, “and we consider sooner or later there will probably be a Copilot for everybody and for every part you do.”
To that finish, the corporate made a slew of copilot-related bulletins and updates at Ignite 2023. For starters, it introduced the final availability of Copilot for Microsoft 365, which it initially unveiled in March.
Since early adopters first began working with Copilot for Microsoft 365, Microsoft has made a number of additions, together with a brand new dashboard that exhibits what the copilot is doing, new personalization capabilities, and new whiteboarding and note-taking capabilities in Copilot for Outlook. Further updates have been added for Copilots for PowerPoint, Excel, and Microsoft Viva.
There’s additionally a brand new Copilot for Service, which is focused at customer support professionals. Safety Copilot, which it launched earlier this yr, will play a distinguished function within the system ensuing from the mix of Sentinel safety analytics and Microsoft Defender XDR platforms.
Copilot for Azure, in the meantime, serves as an AI companion for cloud directors. “Greater than only a software,” Microsoft declares, “it’s a unified chat expertise that understands the person’s function and targets, and enhances the power to design, function and troubleshoot apps and infrastructure.”
The corporate additionally rolled out Copilot Studio, a low-code software designed to permit Microsoft 365 customers to construct their very own customized copilots and join them to enterprise knowledge. Its Bing Chat and Bing Chat Enterprise choices have been changed with (you’ll by no means guess) Copilot. “Once you give Copilot a seat on the desk,” the corporate says, “it goes past being your private assistant to serving to all the group.”
Organizations that use Microsoft Groups to collaborate will quickly have the ability to spin up 3D digital assembly locations utilizing GenAI. Microsoft says its Groups prospects will have the ability to request the creation of 3D conferences and objects utilizing its AI Copilot system. The digital actuality (VR) model of Groups is due in January.
OpenAI and Nvidia Partnerships
Microsoft has an in depth partnership with OpenAI and is invested within the firm. The entire latest new capabilities that OpenAI introduced two weeks in the past at its DevDay occasion–akin to GPT-4 Turbo and GPSs apps–will probably be supplied by Microsoft by way of Azure OpenAI Service too.
“As OpenAI innovates, we’ll ship all of that innovation as a part of Azure OpenAI,” Microsoft CEO Satya Nadella mentioned.
So far as the timeline goes, GPT-3.5 Turbo mannequin with a 16K token immediate size will probably be usually obtainable quickly, and GPT-4 Turbo will probably be obtainable by the tip of the month. GPT-4 Turbo with Imaginative and prescient will quickly be obtainable as a preview.
One other accomplice essential for Microsoft’s ambitions is Nvidia. The GPU chipmaker and the software program large unveiled that its new AI foundry service, which can embrace Nvidia instruments like AI Basis Fashions, NeMo framework, and DGX Cloud AI supercomputing, will probably be obtainable on Azure.
Nvidia CEO Jensen Huang joined Microsoft CEO Nadella on stage. “You invited Nvidia’s ecosystem, all of our software program stacks, to be hosted on Azure,” Huang mentioned. “There’s only a profound transformation in the way in which that Microsoft works with the ecosystem.”
The corporate made a number of bulletins round AI improvement, together with rolling out Azure AI Studio, which the corporate describes as a “hub” for exploring, constructing, testing, and deploying GenAI apps, and even your personal customized copilots.
The corporate additionally unveiled a brand new providing known as Home windows AI Studio that permits developer to construct and run AI fashions straight on the Home windows working system. Home windows AI Studio will permit builders to entry and play with a wide range of language fashions, akin to its personal Microsoft Phi, Meta’s Llama2, and open supply fashions sourced from Azure AI Studio or Hugging Face.
It additionally rolled out Mannequin-as-a-Service, which can give builders entry to the most recent AI fashions from its mannequin catalog. AI builders will have the ability to Llama 2, upcoming premium fashions from Mistral, and Jais from G42, as an API endpoint, the corporate says.
Vector Search, which is a characteristic of Azure AI Search, is now usually obtainable, the corporate says. It additionally added a brand new “immediate stream” functionality to Azure Machine Studying. This may “streamline all the improvement lifecycle” of GenAI and LLM apps, the corporate says.
Microsoft unveiled a brand new Arm-based CPU this week. Dubbed the Azure Cobalt, the brand new chip is 40% sooner than the business Arm chips it at the moment makes use of, the corporate says. The Azure Cobalt will probably be supplied completely within the Azure cloud and is designed for cloud workloads.
It additionally introduced Azure Maia, which it calls an “AI accelerator chip” that’s designed to run cloud-based coaching and inferencing for AI workloads akin to OpenAI fashions, Bing, GitHub Copilot and ChatGPT.
Some Information Stuff Too
It wasn’t all fashions the entire time at Ignite. Information, in any case, lies on the coronary heart of AI, and Microsoft made some data-related bulletins at Ignite.
As an illustration, it introduced that Microsoft Cloth OneLake, which it introduced earlier this yr, is on the market as a knowledge retailer in Azure Machine Studying. The corporate says this may make it simpler for knowledge engineers to share “machine learning-ready knowledge property developed in Cloth.”
Microsoft introduced the GA of Azure Information Lake Storage Gen2 (ADLS Gen2) “shortcuts,” which can permit knowledge engineers “to hook up with knowledge from exterior knowledge lakes in ADLS Gen2 into OneLake by a stay reference to goal knowledge.”
The corporate additionally helps “Amazon S3 shortcuts” in OneLake, which it says will permit prospects to “create a single virtualized knowledge lake” that spans Amazon S3 buckets and OneLake, thereby eliminating the latency concerned with copying knowledge.
You possibly can entry Microsoft’s full slate of AI information from Ignite 2023 right here. The total “e-book ‘o information,” together with all 100 product bulletins made on the present, is obtainable right here.