- Microsoft announced dozens of new products, services, and feature updates at Microsoft Ignite 2023 as it prepares to lead the way in enabling AI development.
- The annual power-packed event is Microsoft’s way of showcasing its premier tools developed as part of its groundwork for its grand strategy.
- At Microsoft Ignite 2023, the tech community got a comprehensive view of Microsoft’s AI ambitions.
The “age of copilots” is upon us. That’s what Microsoft CEO Satya Nadella declared during his keynote address at Microsoft Ignite 2023. The event concluded on Friday alongside the drama unfolding at the Microsoft-backed OpenAI, where the board of directors fired CEO and co-founder Sam Altman, now hired by Nadella.
Microsoft Ignite 2023 is the event’s first edition since the company leaped ahead in generative AI-driven applications and data service offerings. However, the company is leaving no stone unturned in ensuring it continues its dominance into the future by infusing AI into its infrastructure layer as well.
Unsurprisingly, approximately 50% of the ~600 sessions at Microsoft Ignite 2023 were AI-related in some way, as the company brings most of the generative AI products and services under the Copilot umbrella.
“We’re entering this exciting new phase of AI, where we’re not just talking about it as technology that’s new and interesting, but we’re getting into the details of producing, making, deployment, and safety. We’re at a tipping point. This is clearly the age of Copilots,” Nadella said.
From Microsoft’s AI Copilot stack, new AI chips, and more, take a look at Spiceworks News & Insights’ top picks from all announcements made in Redmond this winter.
Microsoft Ignite 2023 Top Announcements
1. Copilot additions
Microsoft introduced four new additions to its Copilot gen AI service offerings suite.
- Copilot for Azure: Introduced for Microsoft’s cloud customer productivity and available within Azure, Copilot for Azure is designed to assist with troubleshooting, problem identification, configuration suggestions, and more.
- Copilot in Dynamics 365 Guides: Aimed at frontline worker productivity, Copilot in Dynamics 365 Guides is created to deliver information at the fingertips. It can be clubbed with mixed reality for various operations such as troubleshooting, training, documentation, guidance, and more.
- Copilot Studio: This is Microsoft, enabling anyone with low code skills to create customized AI assistants. Its 1,000+ pre-built connectors extend data integration plugins for various use cases.
- Copilot for Service: This offering caters to agent productivity engaged in sales and customer service-facing roles, integrating with customer relationship management applications like Salesforce, Zendesk, and ServiceNow.
Microsoft is discontinuing the name Bing Chat. The product will henceforth be known as simply Copilot. The company also added multiple new features in Copilot for Microsoft 365.
2. Windows AI Studio
Windows AI Studio is for developers who want to create AI applications but aren’t keen on leveraging the cloud. This latest tool from Microsoft enables local development, though the obvious compute power constraints will only allow developers to tune and deploy small language models (SLMs).
However, they still have the option to run it on Microsoft Azure, either locally or as a hybrid of the two. Windows AI Studio sources models and other dev tools from Microsoft’s internal Azure AI Studio and external services such as Hugging Face.
Windows AI Studio will be available as a Visual Studio Code extension in the coming weeks.
3. AI chips
Microsoft’s efforts to service the AI-based infrastructure layer include two new internally-made chips: Azure Maia 100 AI Accelerator and the Azure Cobalt 100 CPU.
Both semiconductors should help Microsoft shed its reliance on externally sourced GPUs, including partner NVIDIA’s high-in-demand H100 GPUs. However, both are designated to serve different purposes, with the Azure Maia 100 AI Accelerator intended to train large language models and other generative AI tasks specifically on the Azure infrastructure.
Microsoft is collaborating with OpenAI for feedback on the Maia 100 AI Accelerator, based on the 5 nm process consisting of 105 billion transistors.
Meanwhile, the Arm-based Azure Cobalt 100 CPU is designed for general-purpose computing workloads on the Microsoft Cloud. It is a 64-bit processor with 128 computing cores and is 40% more power efficient than existing Arm-based chips deployed in Azure. Both deliver a throughput of 12.5 gigabytes per second and will be available early in 2024.
“The chips represent a last puzzle piece for Microsoft to deliver infrastructure systems — which include everything from silicon choices, software, and servers to racks and cooling systems — that have been designed from top to bottom and can be optimized with internal and customer workloads in mind,” Microsoft noted.
4. Partnership with NVIDIA
Microsoft and NVIDIA are two of the biggest companies enabling AI development, which have consistently deepened existing partnerships with each other. Even though Microsoft is designing and developing its own AI silicon, NVIDIA may not necessarily perceive it as a threat.
This is evident from the fact that Microsoft is expanding NVIDIA’s foothold in its cloud and data center services, including:
- The introduction of NCC H100 v5 VMs for confidential computing applications
- The Hopper architecture-based H100- and H200-powered virtual machines for Azure
- The availability of Nvidia Omniverse tools on Microsoft’s cloud
- The prevalence and use of NVIDIA supercomputer clusters on Microsoft Azure, which OpenAI uses.
Nvidia co-founder and CEO Jensen Huang called Microsoft “ecosystem partners.” On stage with Nadella, Huang said, “You really deserve so much credit for transforming Microsoft’s entire culture to be so much more collaborative and partner-oriented. You invited Nvidia’s ecosystem, all of our software stacks, to be hosted on Azure.
“And as a result, we’re ecosystem partners,” Huang continued. “There’s just a profound transformation in the way that Microsoft works with the ecosystem.”
5. AI-based cybersecurity offerings
Security Copilot is now available under the Early Access Program to provide data and threat intelligence, malware analysis, incident response, and end-to-end protection across various security products and services.
Microsoft has clubbed two of its premier security offerings — Microsoft Defender XDR and Sentinel — with Security Copilot for a unified security operations experience. Further, Security Copilot features new capabilities to:
- Review and manage data alerts in Microsoft Purview
- Thwart identity compromise with embeddings in Microsoft Entra
- Enable endpoint management with Microsoft Intune
- Deliver guided risk exploration and contextual vulnerability insights with Microsoft Defender for Cloud
- Track asset vulnerabilities with Microsoft Defender for External Attack Surface Management
Microsoft also rolled out multiple other products such as a virtual clone creator (a deepfake tool, in other words), Microsoft Fabric, features such as AI voice isolation and background decoration in Teams, and consolidated To Do, Planner, and Project into one offering.
Image source: Shutterstock