Microsoft Ends Its AI Frontiers Group

John Langford, a machine learning computer scientist formerly at Microsoft Research (MSR), has penned a blog detailing the end of that company’s AI Frontiers Group.

Langford has had a prestigious career in the industry, having worked at the IBM Watson Research Center, Toyota Technological Institute at Chicago, Yahoo Research and, most recently, MSR. According to Langford, the release of ChatGPT shone a spotlight on the shortcomings of MSR’s approach to AI development. As a result, Langford and others formed the AI Frontiers subgroup within the organization.

The new group worked to push AI research further:

In AI Frontiers, we actively tried to organize coherent research efforts. This was not precisely “top down” or “bottom up”, but rather mostly somewhere in the messy middle. There were disagreements of course, and resolving them was sometimes quite challenging. Compute availability was also an often-severe bottleneck. But at a high level I was under the general impression that the organization was succeeding in doing pioneering things, particularly given the small size of AIF (~50 people) and relative to what we could do previously. Much of the work of AIF was figuring out how to make agents work, for example the Autogen orchestration framework and the Fara computer use agent. In NYC however, several of us were interested in developing new fundamental capabilities such as the Next Latent objective which enables extrapolation, Dion3 which creates an optimized/improved Muon, or the Full Bandwidth Transformer which efficiently makes models function substantially more powerfully via recursion.

Unfortunately, a reorganization appears to have been AIF’s undoing:

In early September we were reorged into Microsoft AI. On the surface this could make sense, because putting two different AI organizations together creates opportunities for better concentration of effort, and there were places where AIF was ahead of MAI and of course vice-versa (MAI is much larger). In practice, there are substantial issues because the philosophy of the ex-Inflection leadership of MAI is incompatible. The core philosophy there is scaling, and aggressively not research. This might shock some folks who naturally think that the way you make better models is via research but there is plenty of evidence that bigger models are better models as far as the quality of the predictions they make and from a business perspective returns to scale are far more predictable than returns to research. In a scaling-only environment of course the primary effort is on getting/making/refining data, which is quite a step away from what we were working on. Aside from the incompatible focus, there was plenty of evidence that they just did not really want us.

Over time, the situation became untenable and Langford left:

It’s a tough situation to find yourself in. I consider the scaling-only focus to be both misdirected and not fun. It is an understandable failure mode—scaling is straightforward and it’s common for people to give up on a tradeoff between multiple competing ideals (like innovation and size) by short-sightedly adopting one in totality and dropping the other. But returns to research are real and significant. For example, the Bytedance experiments suggest a variant of the Full Bandwidth Transformer might be comparable to a model of twice the size, implying significant savings in training and inference for the same capability. Do the math—how much is it worth to halve training and inference cost given that the industry has investments approaching a trillion dollars?

So I situated my folks as best I could and left.

Langford’s experience, unfortunately, underscores the experience many have had in the AI field, as well as the tech field in general. Corporate goals and the need to scale technology are often at direct odds with the researchers who originally created or discovered a breakthrough.

Reading Langford’s entire blog post, one can’t help but have a deeper understanding of why the recent AI revolution was sparked within startups, and not inside one of the world’s largest tech companies.