John Langford, the Microsoft researcher who created the company's AI Frontiers group, has left Microsoft and declared the group over, ending a three-year experiment in fundamental artificial intelligence research inside Microsoft Research. Langford announced his departure on social media on Friday, November 6, 2026, and published a blog post titled "The End of AI Frontiers" on his personal site, hunch.net, the same day.

The Microsoft AI Frontiers group had about 50 people, and its best-known work was AutoGen, a widely used open-source framework for building systems of cooperating AI agents. Langford said the group's fate was sealed in early September 2026, when it was moved out of Microsoft Research and into Microsoft AI, a reporting change he said left little room for the kind of research the group was built to do.

Key Points

• John Langford left Microsoft and declared the AI Frontiers research group over on Friday, November 6, 2026.
• Langford created AI Frontiers inside Microsoft Research in 2023; the group had about 50 people.
• In early September 2026, the group was moved into Microsoft AI, a change Langford says left little room for fundamental research.
• The group's public output included the AutoGen multi-agent framework and the Fara computer-use agent.
• An October 16, 2026 change to the AutoGen repository said Microsoft will no longer provide bug fixes, security fixes, new features or support.

What Happened

Langford, a partner research manager at Microsoft Research New York, announced his departure on social media on Friday, November 6, 2026. In his blog post he wrote that the team had tried to organise coherent research efforts, and that the move into Microsoft AI had closed that space down. He described the new leadership's philosophy as "scaling, and aggressively not research" — a phrase that captures the tension at the heart of the group's end.

The reporting change came in early September 2026, about two months before Langford's departure. AI Frontiers was moved out of Microsoft Research, the company's long-standing research arm, and into Microsoft AI, the product-facing organisation. For a group built on open-ended questions, reporting to a unit measured on shipping products meant its agenda would shift to nearer-term goals.

Before leaving, Langford said he placed his team members as best he could, and that there will be no AI Frontiers internships in the coming year. Internships are a principal channel through which doctoral students and early-career researchers join applied AI work, so a year without them closes one entry point into Microsoft's research pipeline.

Background and Context

Langford created AI Frontiers inside Microsoft Research in 2023, assembling a team of about 50 researchers in New York. The group sat in a familiar position for corporate research labs: funded by a company that sells products, but charged with open-ended questions that might not pay off for years.

The September move follows a familiar pattern in large software companies: a group built on open-ended questions finds its agenda reshaped once it reports into a unit closer to shipping products. In Langford's account, the move did not close AI Frontiers outright, but it removed the conditions that made its work possible.

Microsoft AI Frontiers: What the Group Built

AI Frontiers' most visible legacy is AutoGen, an open-source framework for coordinating multiple AI agents — software components that each make model calls and pass work to one another. AutoGen became a widely used tool for building systems in which several agents cooperate, in a field where orchestrating many model calls reliably is still an active problem.

The group's other public project was Fara, an agent that operates a computer to carry out tasks, such as navigating a desktop interface on a user's behalf. Fara was AI Frontiers' attempt to push agents from conversation into action.

Both projects now face an open question about maintenance. An October 16, 2026 change to the AutoGen repository said Microsoft is "not providing bug fixes, security fixes, new features, or support for this project." Open-source frameworks survive their original teams only when someone owns their maintenance, and nothing in Langford's announcement settles who will carry that responsibility for AutoGen or Fara.

Reactions

Langford's own words carry the announcement. In his post he described a leadership philosophy of "scaling, and aggressively not research," and said he had found places for team members as well as he could before leaving.

No statement from Microsoft was included in Langford's announcement. Developers who depend on AutoGen or Fara have good reason to watch for ownership signals in the coming months, since the October repository notice already made clear that Microsoft is stepping back from supporting AutoGen.

Why It Matters

The closure matters first to the developers who build on AutoGen. A widely used framework without a maintainer accumulates unpatched bugs and unanswered security questions, and the October notice means those who relied on Microsoft's backing must now decide whether to fork the project, find another maintainer, or move on.

It also matters as a signal about where Microsoft's AI effort is heading. The end of a research group whose leader says its new home had no room for fundamental research suggests that, inside Microsoft, the balance has tilted decisively toward scaling what already works. The group's ideas do not vanish — its people and methods move elsewhere — but the organisational home for that kind of work at Microsoft is now narrower.

What Happens Next

The immediate question is ownership. Watch for signals about who, if anyone, takes responsibility for AutoGen and Fara — another Microsoft team, the community, or no one at all. The October notice points toward a wind-down unless someone steps in.

Langford has not said what he will do next, and the former AI Frontiers researchers are now scattered across other teams and institutions. There will be no AI Frontiers internships in the coming year. Whether the group's ideas find a durable home elsewhere is still open.

Frequently Asked Questions

What was Microsoft AI Frontiers?
AI Frontiers was a research group inside Microsoft Research, created in 2023 by John Langford, with about 50 people. It worked on fundamental AI research, including the AutoGen framework and the Fara computer-use agent.

Why did John Langford leave Microsoft?
Langford said the group was moved out of Microsoft Research and into Microsoft AI in early September 2026, and that the new leadership's philosophy was "scaling, and aggressively not research," leaving little room for fundamental research.

What is AutoGen?
AutoGen is an open-source framework for coordinating multiple AI agents, software components that each make model calls and pass work to one another. It is widely used for building multi-agent systems.

Is AutoGen still supported?
No. An October 16, 2026 change to the AutoGen repository said Microsoft is "not providing bug fixes, security fixes, new features, or support for this project."

What happens to the group's researchers?
Langford said he placed his team members as best he could before leaving. There will be no AI Frontiers internships in the coming year.

Read more on US News Pulse: Meta's $95 billion AI raise leads a $198 billion day of AI spending, Anthropic files for IPO as Claude maker seeks Nasdaq listing and Claude AI bypassed a paywall to reach Georgia vote counts.