Starting next month, Microsoft will allow customers to build autonomous AI agents, marking another step in its efforts to capitalize on the rapidly growing AI technology sector amid increased investor scrutiny of its significant AI investments.
Unlike chatbots, these autonomous agents require minimal human intervention and are being promoted by Microsoft as “apps for an AI-driven world,” designed to handle tasks such as customer queries, identifying sales leads, and managing inventory. Other tech companies, including Salesforce, are also emphasizing the potential of these agents, which some analysts believe could provide a more straightforward path for companies to monetize their substantial AI investments.
Read more: Global IT outage disrupts Microsoft machines, including those in Pakistan
Microsoft customers will be able to use Copilot Studio, a low-code application, to create these agents in a public preview available from November. The agents are powered by a combination of AI models developed in-house and by OpenAI. Additionally, Microsoft is rolling out 10 pre-built agents that can handle routine tasks, including supply chain management, expense tracking, and client communication.
In a demonstration, McKinsey & Co., with early access to the tools, created an AI agent capable of managing client inquiries, checking interaction history, identifying the appropriate consultant, and scheduling follow-up meetings.
Microsoft’s vision, as explained by Charles Lamanna, corporate VP of Copilot, is that every employee will have a personalized AI agent to interact with other AI agents. However, despite these advancements, there are concerns about the slow adoption of Copilot, with a recent Gartner survey indicating that most companies have yet to move beyond the pilot stage.
