Introducing Microsoft Research Focus, a new way to catch up on research news from across Microsoft.

In this edition, we're covering advances in AI agents, what patterns in everyday AI usage reveal about how people work, BioEmu2's public preview, and research aimed at improving in advance.

Watch the full video here or listen on major podcast platforms.

Think everyone else understands except you?

Microsoft Partner Research Manager Jennifer Neville shares how she learned that curiosity sometimes requires courage—and that someone else probably has the same questions as you.

Check out the full conversation. msft.it/6016a0wcC

The first intern at Microsoft Research Asia–Tokyo arrived as a computer vision researcher. She left with a broader view of physical intelligence, spanning perception, language, learning, and real-world action. msft.it/6011a0uOb

Can AI agents solve complex problems more efficiently? Plus, tackling coding bottlenecks; evolving agent skills from telemetry; how AI needs vary across industries; an AI tool saving teachers time; a weather forecast breakthrough; BioEmu2 in public preview. msft.it/6017aI27v

Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. msft.it/6016aaFOu

AI agents are writing more code. How can developers ensure that code is correct?

Microsoft Research’s Proof-Driven Development (PDD) Kit explores a more rigorous approach: it translates a developer’s intent into a formal specification, then uses formal verification to prove that .

PDD Kit works with Verus for Rust and is integrated with @GitHubNext’s Chopin, bringing formal methods into an AI-assisted development workflow for teams.

See the work in action at @github Universe, Oct. 28–29, 2026.

Join @githubuniverse Oct 28–29 in SF or online. Explore AI, security, and open source with hands-on sessions, product updates, and devs from around the world. msft.it/6015aa5OS

For decades, a key question about the challenges of high-dimensional computation resisted proof. New research resolves the KLS conjecture, removing a major barrier to understanding the efficiency of randomized algorithms used throughout science and computing. o

What keeps you hopeful when the problems you're working to solve feel overwhelming?

For Microsoft Research's Chris White, it's knowing there's something he can do.

Catch the latest episode of the Microsoft Research Podcast. msft.it/6014av214

Last week at the OCP Global Summit, @MediaTek demoed a live MicroLED optical link in a standard pluggable form factor, built on technology developed at Microsoft Research: inexpensive LEDs instead of copper or lasers to send many parallel streams of data at low power. This is a meaningful next step toward AI datacenters that are more reliable and use less power.

Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security: msft.it/6016atJb6

Weather forecasts are unreliable past two weeks, leaving a crucial gap for farmers and others. Probabilistic Bias Correction improves extended forecasting, taking top honors in a global competition and showing promise for real-world use: msft.it/6013aseY9

IMAGE CREDIT: or Research in the Atmosphere at Colorado State University and the National Oceanic and Atmospheric Administration (CSU/CIRA & NOAA)

AI does better when it acknowledges it needs clarification; Improving robot performance on vision-language tasks; How we learn from AI’s failures; Improving AI tutoring; Understanding our sense of smell; More efficient KV-cache memory storage msft.it/6019aSVIP

StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics: msft.it/6013aS3JF

The most important AI failures may not be the obvious ones.

Microsoft Partner Research Manager Jennifer Neville explains to host Chad Atalla why “surprising failures” can reveal where human expectations about intelligence diverge from how AI systems actually work. msft.it/6017aU8r7

Training AI agents with reinforcement learning can be challenging because their tools, context, and decision-making are managed by complex frameworks. Agent Lightning connects existing agents to RL training, making it easier to improve them without rebuilding them.: msft.it/6019aokV1

Progress can be a winding road.

Jennifer Neville discusses spinning her research wheels, taking a multiturn path to computer science, and pushing today’s AI systems with more practical evaluations.

Catch the latest Microsoft Research Podcast episode.: msft.it/6183aW0lD

Extreme space-weather events can damage power systems on Earth and degrade GPS accuracy and satellite operations. A new machine learning system can predict where damage is likely to occur 30-60 minutes before a storm arrives: msft.it/6019aYXwh

Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. By connecting insights across biological scales and modalities, Quine helps scientists computationally search a ntuition allows and prioritize hypotheses before they reach the lab. Experimental results provide important feedback, helping researchers sharpen future research directions. msft.it/6014acaRn

Since launching a year ago, the Microsoft Research Asia — Singapore lab has established a strong foundation, deepened collaboration across government, academia, and industry, and explored how frontier AI research can create real-world value: msft.it/6019acwE5