Tag: Microsoft

Categories
GeekWire

How AI can help scientists head off water woes

Screenshot showing interactive map generated in response to AI query: "Show me the drought condition in North America for August 2012"
A screenshot shows how Hydrology Copilot can turn a simple query into an interactive, color-coded map. (Microsoft Infographic)

Microsoft and NASA say they’re applying artificial intelligence to a challenge that has become increasingly urgent: how to cope with flooding and other disasters driven by extreme weather.

The result of their efforts is Hydrology Copilot, a set of AI agents aimed at making hydrological data easier to access and analyze. The platform is built on the foundation that was established for NASA Earth Copilot, a cloud-based AI tool that can sift through petabytes of Earth science data.

Hydrology is the scientific study of Earth’s water cycle, which encompasses precipitation, runoff, evaporation and the movement of water through rivers, lakes and soil. It’s not just an academic exercise: Hydrologic insights are put to use in fields ranging from agriculture to forestry to urban development.

“NASA has long produced advanced hydrology and land-surface datasets, powering breakthroughs in drought early-warning systems, environmental planning and environmental research,” Juan Carlos López, a senior solution specialist at Microsoft who focuses on space and AI, wrote in a blog post. “Yet despite their value, these datasets and the specialized tools required to navigate and interpret them remain difficult to access for many who could benefit most.”

That’s where Hydrology Copilot comes in: Powered by Microsoft Azure OpenAI Service and Microsoft Foundry, the platform lets researchers query NASA’s data using straightforward questions — for example, “Which regions may be facing elevated flood risk?”

Get the full story on GeekWire
Categories
GeekWire

Prediction: Human-level AI is just a few ‘miracles’ away

Nathan Myhrvold and Todd Bishop on stage at Microsoft@50 event
Intellectual Ventures CEO Nathan Myhrvold makes a point during a chat with GeekWire's Todd Bishop. (GeekWire Photo / Kevin Lisota)

Will artificial intelligence ever catch up with human intelligence? And if it does, is humanity doomed? Intellectual Ventures CEO Nathan Myhrvold, who had the job of predicting the future of tech during Microsoft’s early years, was ready with some answers this week at GeekWire’s Microsoft@50 anniversary event.

The feature that AI still lacks is the ability to create a new abstract concept, “imbue it with meaning and then reason about it,” said Myhrvold, who joined Microsoft in 1986 and served as the company’s first chief technology officer.

“I think we’ll get there, but that’s at least one miracle that needs to be figured out, and I variously have thought there was like three to five miracles that need to be done. Who knows?” he said during the March 20 event at Town Hall Seattle. “And that could happen tomorrow, or maybe it already happened tonight, and they just haven’t told us. Or it could take another 10 years.”

GeekWire co-founder Todd Bishop returned to the topic a minute later. “Did I hear correctly that we’re three to five miracles away from AI that’s as powerful or as intelligent as humans?” he asked.

“Yes,” Myhrvold replied.

Get the full story on GeekWire
Categories
GeekWire

Microsoft and Atom push ahead on the quantum frontier

Atom Computing quantum computer in server room
Atom Computing is working with Microsoft on fault-tolerant quantum computers. (Atom Computing Photo)

Microsoft and Atom Computing say they’ve reached a new milestone in their effort to build fault-tolerant quantum computers that can show an advantage over classical computers.

Microsoft says it will start delivering the computers’ quantum capabilities to customers by the end of 2025, with availability via the Azure cloud service as well as through on-premises hardware.

“Together, we are co-designing and building what we believe will be the world’s most powerful quantum machine,” Jason Zander, executive vice president at Microsoft, said in a LinkedIn posting.

Like other players in the field, Microsoft’s Azure Quantum team and Atom Computing aim to capitalize on the properties of quantum systems — where quantum bits, also known as qubits, can process multiple values simultaneously. That’s in contrast to classical systems, which typically process ones and zeros to solve algorithms.

Microsoft has been working with Colorado-based Atom Computing on hardware that uses the nuclear spin properties of neutral ytterbium atoms to run quantum calculations. One of the big challenges is to create a system that can correct the errors that turn up during the calculations due to quantum noise. The solution typically involves knitting together “physical qubits” to produce an array of “logical qubits” that can correct themselves.

In a paper posted to the ArXiv preprint server, members of the research team say they were able to connect 256 noisy neutral-atom qubits using Microsoft’s qubit-virtualization system in such a way as to produce a system with 24 logical qubits.

“This represents the highest number of entangled logical qubits on record,” study co-author Krysta Svore, vice president of advanced quantum development for Microsoft Azure Quantum, said today in a blog posting. “Entanglement of the qubits is evidenced by their error rates being significantly below the 50% threshold for entanglement.”

Get the full story on GeekWire
Categories
GeekWire

NASA and Microsoft team up on AI guide to Earth data

Screenshot of Earth Copilot interface with image of hurricane
A screenshot shows Earth Copilot in action, guiding a computer user to data about Hurricane Harvey. (Image via Microsoft)

Microsoft and NASA have joined forces to develop Earth Copilot, a software tool that uses artificial intelligence and cloud computing to simplify the process of searching and analyzing Earth science data.

The collaboration is part of NASA’s Transform to Open Science Initiative, which aims to make more than 100 petabytes of satellite imagery and other Earth science data more readily available to scientists, educators, policymakers and the general public.

“The vision behind this collaboration was to leverage AI and cloud technologies to bring Earth’s insights to communities that have been underserved, where access to data can lead to tangible improvements,” Minh Nguyen, Cloud Solution Architect at Microsoft, said today in a blog posting about the project. “By enabling users to interact with the data through simple, plain language queries, we’re helping to democratize access to spaceborne information.”

Tyler Bryson, Microsoft’s corporate vice president for U.S. health and public sector industries, said Earth Copilot lets users interact with NASA’s data repository conversationally.

Bryson noted that the sheer scale and complexity of the information gathered for NASA’s Earth Science Data Systems Program can sometimes make finding the right information a daunting task. Earth Copilot doesn’t require users to navigate highly technical interfaces or master the intricacies of geospatial analysis.

“Instead, they can simply ask questions such as ‘What was the impact of Hurricane Ian in Sanibel Island?’ or ‘How did the COVID-19 pandemic affect air quality in the U.S.?’” Bryson said. “AI will then retrieve relevant datasets, making the process seamless and intuitive.”

Get the full story on GeekWire
Categories
GeekWire

Microsoft works with students on Golden Record 2.0

Richard Black holding glass platter
Richard Black, research director for Microsoft’s Project Silica, holds a data storage platter. (Microsoft Photo)

Forty-seven years after NASA sent a “Golden Record” into deep space to document humanity’s view of the world, Microsoft’s Project Silica is teaming up with a citizen-science effort to lay the groundwork — or, more aptly, the glasswork — for doing something similar.

Golden Record 2.0, a project created by students, teachers and researchers affiliated with Avenues: The World School, is also getting an assist from artist Jon Lomberg, who was the design director for Golden Record 1.0.

The original Golden Record project involved preserving imagery and sounds from around the world on gold-plated phonograph records. Copies of the record were placed on NASA’s Voyager 1 and 2 probes and launched into space in 1977. The idea was that if space travelers came across the records in the distant future, they could decipher the recorded archive and learn what our world was like in the 20th century.

Golden Record 2.0’s organizers are going after the same idea, even though they’re still looking into how their archive would be packaged and launched.

Project Silica could play a role in the packaging. Richard Black, a manager at Microsoft Research’s Cambridge lab in Britain, has been leading an effort to store data inside thin platters of fused silica glass.

“It does that using ultrashort laser pulses that make a permanent, detectable and yet transparent modification to the glass crystal, so the data ends up as durable as the piece of glass itself,” Black explained in a Microsoft podcast called Collaborators.

Each coaster-sized platter could store several terabytes of data for 10,000 years or more, according to Microsoft. The data can be read out using a microscope, and decoded using machine-learning algorithms.

Get the full story on GeekWire
Categories
GeekWire

Scientists harness generative AI for cancer diagnosis

Illustration: Visualization of microchip with the look of a whole-slide image
Prov-GigaPath harnesses AI for digital pathology. (Ella Maru Studio via Providence / Microsoft)

Researchers at Microsoft, Providence Health System and the University of Washington say they’ve developed a new artificial intelligence model for diagnosing cancer, based on an analysis of more than a billion images of tissue samples from more than 30,000 patients.

The open-access model, known as Prov-GigaPath, is described in research published today by the journal Nature and is already being used in clinical applications.

“The rich data in pathology slides can, through AI tools like Prov-GigaPath, uncover novel relationships and insights that go beyond what the human eye can discern,” study co-author Carlo Bifulco, chief medical officer of Providence Genomics, said in a news release. “Recognizing the potential of this model to significantly advance cancer research and diagnostics, we felt strongly about making it widely available to benefit patients globally. It’s an honor to be part of this groundbreaking work.”

The effort to develop Prov-GigaPath used AI tools to identify patterns in 1.3 billion pathology image tiles obtained from 171,189 digital whole-slides provided by Providence. The researchers say this was the largest pre-training effort to date with whole-slide modeling — drawing upon a database five to 10 times larger than datasets such as the The Cancer Genome Atlas.

Get the full story on GeekWire
Categories
GeekWire

How to turn down the noise in quantum computing

Quantum researchers in lab
Scientists make adjustments to a beam-line array used to deliver laser pulses in quantum computers. (Quantinuum Photo)

Microsoft and Quantinuum say they’ve demonstrated a quantum computing system that can reduce the error rate for data processing by a factor of 800.

“Today signifies a major achievement for the entire quantum ecosystem,” Jason Zander, Microsoft’s executive vice president for strategic missions and technologies, said in a blog posting about the achievement.

Quantum computing could solve certain types of problems — ranging from data encryption and system optimization to the development of new synthetic materials — on a time scale that would be unachievable using classical computers. “Scaled quantum computers would offer the ability to simulate the interactions of molecules and atoms at the quantum level beyond the reach of classical computers, unlocking solutions that can be a catalyst for positive change in our world,” Zander said.

The secret to success lies in quantum bits, or qubits, that can represent multiple values until the results of a computation are read out. Qubits typically make use of exotic materials, such as superconducting circuits, diamonds with defects or laser-cooled ions.

One big challenge is that qubits tend to be “noisy” — that is susceptible to perturbations that introduce errors. For years, researchers have been hunting for ways to maintain the fidelity of qubits and correct any errors that arise. Such strategies typically involve linking up multiple physical qubits to represent a single “logical qubit.”

Just a couple of years ago, Microsoft researchers were saying that a quantum computer would need at least a million physical qubits in order to demonstrate an advantage over classical computers. But that’s because it was thought that thousands of physical qubits would be required to produce a single logical qubit. If fewer physical qubits are required for error correction, that would make it easier to build useful quantum computers.

The newly reported demonstration addresses that challenge: Microsoft and Quantinuum said they created four highly reliable logical qubits from just 30 physical qubits. “With this system, we ran more than 14,000 individual experiments without an error,” Zander said.

Get the full story on GeekWire
Categories
GeekWire

How satellites and AI work together to monitor the planet

Screengrab of global imagery map from Satlas
Satlas’ AI modeling software improves satellite image resolution by a factor of four. (AI2 / Satlas Image)

Geospatial data analysis promises to revolutionize the way agriculture, urban planning and disaster relief is done — and thanks to a variety of projects that make use of artificial intelligence, Microsoft and Seattle’s Allen Institute for AI are part of that revolution.

The Allen Institute for AI, also known as AI2, recently rolled out Satlas, a new software platform for exploring global geospatial data generated from satellite imagery. Meanwhile, Microsoft’s AI for Good Lab is working with public and private institutions in Colombia on Project Guacamaya, which uses AI tools to monitor and understand conditions in the Amazon Rainforest.

Get the full story on GeekWire
Categories
GeekWire

Could the Pacific Northwest become ‘Quantum Valley’?

Physicists adjust quantum hardware at PNNL
Christian Boutan and Jihee Yang adjust quantum hardware at Pacific Northwest National Laboratory. (PNNL Photo / Andrea Starr)

California’s Silicon Valley may rule the roost for internet startups, and Kendall Square in the Boston area may set the pace for the biotech industry, but could the Pacific Northwest leap into the lead for quantum computing?

Experts who are exploring one of the computing world’s hottest frontiers say there’s a chance that Seattle could become the heart of a “Quantum Valley” — but it could take years for the promise of quantum computing to pay off.

“I’d say we’re on the order of a decade plus before we start to see this start to really materialize,” said Louis Terminello, associate laboratory director for physical and computational sciences at the U.S. Department of Energy’s Pacific Northwest National Laboratory.

Terminello is among the authors of a recent commentary calling for increased public investment to solidify the Pacific Northwest as home of the nation’s top regional cluster for research and development in quantum information science — in short, Quantum Valley.

Get the full story on GeekWire
Categories
GeekWire

Azure Quantum joins the hunt for new fuel-cell catalysts

A hydrogen-powered car fills up at a refueling station
A hydrogen-powered car fills up at a refueling station. (Johnson Matthey via YouTube)

Chemists at Microsoft Azure Quantum are teaming up with Johnson Matthey, a British-based clean-tech company, to identify new types of catalysts for hydrogen fuel cells.

The project demonstrates how quantum information science could help reduce the automobile industry’s carbon footprint and address the challenge of climate change.

“So far, Johnson Matthey has seen a twofold acceleration in quantum chemistry calculations, and we’re just getting started,” Nathan Baker, senior director of partnerships for chemistry and materials at Microsoft, said today in a blog posting. “Both companies recognize that the discoveries needed to create a zero-carbon future will require significant breakthroughs in chemical and materials science, and are enthusiastic about the difference we can make in the world together.”

Get the full story on GeekWire