John Langford leaves Microsoft, says AI Frontiers is over after reorg into scaling-focused Microsoft AI

Nov 8, 2026
Langford wrote that AI Frontiers' September move into Microsoft AI clashed with a scaling-only philosophy, and that the group is over.

Machine learning researcher John Langford has left Microsoft and says its AI Frontiers group is over, he wrote on his blog. AI Frontiers, a roughly 50-person group set up within Microsoft Research in 2023, worked on making agents work, including the Autogen orchestration framework and the Fara computer use agent.

In early September the group was moved into Microsoft AI. Langford wrote that the core philosophy of Microsoft AI's ex-Inflection leadership "is scaling, and aggressively not research," and called the scaling-only focus "both misdirected and not fun." He said he could not get innovation into use, or even seriously considered, while in Microsoft AI.

"So I situated my folks as best I could and left," Langford wrote, adding that every other founding member of the group's leadership team seems to have reached a similar conclusion. He decided against returning to Microsoft Research, saying he was not comfortable at a company where his research has no path to real use, and that there will be no internships with the group in the coming year.

The departure points to a wider split inside Microsoft over research priorities, The420 reported.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.