Microsoft CEO Satya Nadella, during the Microsoft Ignite conference on Tuesday, revealed that the company is currently "teaching" new AI tools that would have the capacity to act on behalf of humans in both work and life.
Developers of AI are looking at the next wave of AI chatbots as "agents" that can do more things for people. However, one setback in the development of such tools is the high cost.
On a blog on Tuesday, Microsoft elaborated on the benefits of AI agents for companies, highlighting how it can help businesses to accomplish more.
One example that the tech giant gave was on handling shipping and returns. AI agents "can operate around the clock to review and approve customer returns or go over shipping invoices to help businesses avoid costly supply-chain errors."
It also added that "they can reason over reams of product information to give field technicians step-by-step instructions or use context and memory to open and close tickets for an IT help desk."
Jared Spataro, the chief marketing officer of Microsoft's AI at Work, said that one must regard agents as "the new apps for an AI-powered world."
He also emphasized that they are adding new capabilities that would be a solution to some of the biggest challenges that people face at work and thereafter provide real business results.
OpenAI's recently announced o1 series, can bring more advanced reasoning capabilities to agents, allowing them to take on more complicated tasks by breaking them down into steps such as getting the information someone on an IT help desk would need to solve a problem, factoring in solutions they've tried and coming up with a plan.
Just last month, Microsoft made a pronouncement that it was preparing the world where "every organization will have a constellation of agents — ranging from simple prompt-and-response to fully autonomous," the Associated Press reported.
The annual Ignite conference of Microsoft caters to its huge business customers. Many users have started noticing the limitations of chatbots like ChatGPT, Gemini, and Copilot, which work by predicting the most plausible next word in sentences. This slowly ushered the shift towards agentic AI, which is said to work better in longer-range planning and decision making. This aspect allows these agents to control computers and perform tasks on behalf of humans.
Marc Benioff, the CEO of Salesforce, expressed doubt on the move of Microsoft, calling the re-branding of the giant's Copilot into "agents" as "panic mode." Benioff stated that Copilot was actually a "flop," claiming that the assistant was inaccurate.
On the other hand, Ece Kamar, the managing director of Microsoft's AI Frontiers Lab, put forward positive thoughts on agentic AI.
"If you want to have a system that can really solve real world problems and help people, that system has to have a good understanding of the world we live in, and when something happens, that system has to perceive that change and take action accordingly," he said.
