Google DeepMind’s new AI uses large language models to crack real-world problems

Stay Ahead, Stay ONMINE

Google DeepMind’s new AI uses large language models to crack real-world problems

Google DeepMind has once again used large language models to discover new solutions to long-standing problems in math and computer science. This time the firm has shown that its approach can not only tackle unsolved theoretical puzzles, but improve a range of important real-world processes as well. Google DeepMind’s new tool, called AlphaEvolve, uses the Gemini 2.0 family of large language models (LLMs) to produce code for a wide range of different tasks. LLMs are known to be hit and miss at coding. The twist here is that AlphaEvolve scores each of Gemini’s suggestions, throwing out the bad and tweaking the good, in an iterative process, until it has produced the best algorithm it can. In many cases, the results are more efficient or more accurate than the best existing (human-written) solutions. “You can see it as a sort of super coding agent,” says Pushmeet Kohli, a vice president at Google DeepMind who leads its AI for Science teams. “It doesn’t just propose a piece of code or an edit, it actually produces a result that maybe nobody was aware of.” In particular, AlphaEvolve came up with a way to improve the software Google uses to allocate jobs to its many millions of servers around the world. Google DeepMind claims the company has been using this new software across all of its data centers for more than a year, freeing up 0.7% of Google’s total computing resources. That might not sound like much, but at Google’s scale it’s huge. Jakob Moosbauer, a mathematician at the University of Warwick in the UK, is impressed. He says the way AlphaEvolve searches for algorithms that produce specific solutions—rather than searching for the solutions themselves—makes it especially powerful. “It makes the approach applicable to such a wide range of problems,” he says. “AI is becoming a tool that will be essential in mathematics and computer science.” AlphaEvolve continues a line of work that Google DeepMind has been pursuing for years. Its vision is that AI can help to advance human knowledge across math and science. In 2022, it developed AlphaTensor, a model that found a faster way to solve matrix multiplications—a fundamental problem in computer science—beating a record that had stood for more than 50 years. In 2023, it revealed AlphaDev, which discovered faster ways to perform a number of basic calculations performed by computers trillions of times a day. AlphaTensor and AlphaDev both turn math problems into a kind of game, then search for a winning series of moves. FunSearch, which arrived in late 2023, swapped out game-playing AI and replaced it with LLMs that can generate code. Because LLMs can carry out a range of tasks, FunSearch can take on a wider variety of problems than its predecessors, which were trained to play just one type of game. The tool was used to crack a famous unsolved problem in pure mathematics. AlphaEvolve is the next generation of FunSearch. Instead of coming up with short snippets of code to solve a specific problem, as FunSearch did, it can produce programs that are hundreds of lines long. This makes it applicable to a much wider variety of problems. In theory, AlphaEvolve could be applied to any problem that can be described in code and that has solutions that can be evaluated by a computer. “Algorithms run the world around us, so the impact of that is huge,” says Matej Balog, a researcher at Google DeepMind who leads the algorithm discovery team. Survival of the fittest Here’s how it works: AlphaEvolve can be prompted like any LLM. Give it a description of the problem and any extra hints you want, such as previous solutions, and AlphaEvolve will get Gemini 2.0 Flash (the smallest, fastest version of Google DeepMind’s flagship LLM) to generate multiple blocks of code to solve the problem. It then takes these candidate solutions, runs them to see how accurate or efficient they are, and scores them according to a range of relevant metrics. Does this code produce the correct result? Does it run faster than previous solutions? And so on. AlphaEvolve then takes the best of the current batch of solutions and asks Gemini to improve them. Sometimes AlphaEvolve will throw a previous solution back into the mix to prevent Gemini from hitting a dead end. When it gets stuck, AlphaEvolve can also call on Gemini 2.0 Pro, the most powerful of Google DeepMind’s LLMs. The idea is to generate many solutions with the faster Flash but add solutions from the slower Pro when needed. These rounds of generation, scoring, and regeneration continue until Gemini fails to come up with anything better than what it already has. Number games The team tested AlphaEvolve on a range of different problems. For example, they looked at matrix multiplication again to see how a general-purpose tool like AlphaEvolve compared to the specialized AlphaTensor. Matrices are grids of numbers. Matrix multiplication is a basic computation that underpins many applications, from AI to computer graphics, yet nobody knows the fastest way to do it. “It’s kind of unbelievable that it’s still an open question,” says Balog. The team gave AlphaEvolve a description of the problem and an example of a standard algorithm for solving it. The tool not only produced new algorithms that could calculate 14 different sizes of matrix faster than any existing approach, it also improved on AlphaTensor’s record-beating result for multipying two four-by-four matrices. AlphaEvolve scored 16,000 candidates suggested by Gemini to find the winning solution, but that’s still more efficient than AlphaTensor, says Balog. AlphaTensor’s solution also only worked when a matrix was filled with 0s and 1s. AlphaEvolve solves the problem with other numbers too. “The result on matrix multiplication is very impressive,” says Moosbauer. “This new algorithm has the potential to speed up computations in practice.” Manuel Kauers, a mathematician at Johannes Kepler University in Linz, Austria, agrees: “The improvement for matrices is likely to have practical relevance.” By coincidence, Kauers and a colleague have just used a different computational technique to find some of the speedups AlphaEvolve came up with. The pair posted a paper online reporting their results last week. “It is great to see that we are moving forward with the understanding of matrix multiplication,” says Kauers. “Every technique that helps is a welcome contribution to this effort.” Real-world problems Matrix multiplication was just one breakthrough. In total, Google DeepMind tested AlphaEvolve on more than 50 different types of well-known math puzzles, including problems in Fourier analysis (the math behind data compression, essential to applications such as video streaming), the minimum overlap problem (an open problem in number theory proposed by mathematician Paul Erdős in 1955), and kissing numbers (a problem introduced by Isaac Newton that has applications in materials science, chemistry, and cryptography). AlphaEvolve matched the best existing solutions in 75% of cases and found better solutions in 20% of cases. Google DeepMind then applied AlphaEvolve to a handful of real-world problems. As well as coming up with a more efficient algorithm for managing computational resources across data centers, the tool found a way to reduce the power consumption of Google’s specialized tensor processing unit chips. AlphaEvolve even found a way to speed up the training of Gemini itself, by producing a more efficient algorithm for managing a certain type of computation used in the training process. Google DeepMind plans to continue exploring potential applications of its tool. One limitation is that AlphaEvolve can’t be used for problems with solutions that need to be scored by a person, such as lab experiments that are subject to interpretation. Moosbauer also points out that while AlphaEvolve may produce impressive new results across a wide range of problems, it gives little theoretical insight into how it arrived at those solutions. That’s a drawback when it comes to advancing human understanding. Even so, tools like AlphaEvolve are set to change the way researchers work. “I don’t think we are finished,” says Kohli. “There is much further that we can go in terms of how powerful this type of approach is.”

Google DeepMind’s new tool, called AlphaEvolve, uses the Gemini 2.0 family of large language models (LLMs) to produce code for a wide range of different tasks. LLMs are known to be hit and miss at coding. The twist here is that AlphaEvolve scores each of Gemini’s suggestions, throwing out the bad and tweaking the good, in an iterative process, until it has produced the best algorithm it can. In many cases, the results are more efficient or more accurate than the best existing (human-written) solutions.

“You can see it as a sort of super coding agent,” says Pushmeet Kohli, a vice president at Google DeepMind who leads its AI for Science teams. “It doesn’t just propose a piece of code or an edit, it actually produces a result that maybe nobody was aware of.”

In particular, AlphaEvolve came up with a way to improve the software Google uses to allocate jobs to its many millions of servers around the world. Google DeepMind claims the company has been using this new software across all of its data centers for more than a year, freeing up 0.7% of Google’s total computing resources. That might not sound like much, but at Google’s scale it’s huge.

Jakob Moosbauer, a mathematician at the University of Warwick in the UK, is impressed. He says the way AlphaEvolve searches for algorithms that produce specific solutions—rather than searching for the solutions themselves—makes it especially powerful. “It makes the approach applicable to such a wide range of problems,” he says. “AI is becoming a tool that will be essential in mathematics and computer science.”

AlphaEvolve continues a line of work that Google DeepMind has been pursuing for years. Its vision is that AI can help to advance human knowledge across math and science. In 2022, it developed AlphaTensor, a model that found a faster way to solve matrix multiplications—a fundamental problem in computer science—beating a record that had stood for more than 50 years. In 2023, it revealed AlphaDev, which discovered faster ways to perform a number of basic calculations performed by computers trillions of times a day. AlphaTensor and AlphaDev both turn math problems into a kind of game, then search for a winning series of moves.

FunSearch, which arrived in late 2023, swapped out game-playing AI and replaced it with LLMs that can generate code. Because LLMs can carry out a range of tasks, FunSearch can take on a wider variety of problems than its predecessors, which were trained to play just one type of game. The tool was used to crack a famous unsolved problem in pure mathematics.

AlphaEvolve is the next generation of FunSearch. Instead of coming up with short snippets of code to solve a specific problem, as FunSearch did, it can produce programs that are hundreds of lines long. This makes it applicable to a much wider variety of problems.

In theory, AlphaEvolve could be applied to any problem that can be described in code and that has solutions that can be evaluated by a computer. “Algorithms run the world around us, so the impact of that is huge,” says Matej Balog, a researcher at Google DeepMind who leads the algorithm discovery team.

Survival of the fittest

Here’s how it works: AlphaEvolve can be prompted like any LLM. Give it a description of the problem and any extra hints you want, such as previous solutions, and AlphaEvolve will get Gemini 2.0 Flash (the smallest, fastest version of Google DeepMind’s flagship LLM) to generate multiple blocks of code to solve the problem.

It then takes these candidate solutions, runs them to see how accurate or efficient they are, and scores them according to a range of relevant metrics. Does this code produce the correct result? Does it run faster than previous solutions? And so on.

AlphaEvolve then takes the best of the current batch of solutions and asks Gemini to improve them. Sometimes AlphaEvolve will throw a previous solution back into the mix to prevent Gemini from hitting a dead end.

When it gets stuck, AlphaEvolve can also call on Gemini 2.0 Pro, the most powerful of Google DeepMind’s LLMs. The idea is to generate many solutions with the faster Flash but add solutions from the slower Pro when needed.

These rounds of generation, scoring, and regeneration continue until Gemini fails to come up with anything better than what it already has.

Number games

The team tested AlphaEvolve on a range of different problems. For example, they looked at matrix multiplication again to see how a general-purpose tool like AlphaEvolve compared to the specialized AlphaTensor. Matrices are grids of numbers. Matrix multiplication is a basic computation that underpins many applications, from AI to computer graphics, yet nobody knows the fastest way to do it. “It’s kind of unbelievable that it’s still an open question,” says Balog.

The team gave AlphaEvolve a description of the problem and an example of a standard algorithm for solving it. The tool not only produced new algorithms that could calculate 14 different sizes of matrix faster than any existing approach, it also improved on AlphaTensor’s record-beating result for multipying two four-by-four matrices.

AlphaEvolve scored 16,000 candidates suggested by Gemini to find the winning solution, but that’s still more efficient than AlphaTensor, says Balog. AlphaTensor’s solution also only worked when a matrix was filled with 0s and 1s. AlphaEvolve solves the problem with other numbers too.

“The result on matrix multiplication is very impressive,” says Moosbauer. “This new algorithm has the potential to speed up computations in practice.”

Manuel Kauers, a mathematician at Johannes Kepler University in Linz, Austria, agrees: “The improvement for matrices is likely to have practical relevance.”

By coincidence, Kauers and a colleague have just used a different computational technique to find some of the speedups AlphaEvolve came up with. The pair posted a paper online reporting their results last week.

“It is great to see that we are moving forward with the understanding of matrix multiplication,” says Kauers. “Every technique that helps is a welcome contribution to this effort.”

Real-world problems

Matrix multiplication was just one breakthrough. In total, Google DeepMind tested AlphaEvolve on more than 50 different types of well-known math puzzles, including problems in Fourier analysis (the math behind data compression, essential to applications such as video streaming), the minimum overlap problem (an open problem in number theory proposed by mathematician Paul Erdős in 1955), and kissing numbers (a problem introduced by Isaac Newton that has applications in materials science, chemistry, and cryptography). AlphaEvolve matched the best existing solutions in 75% of cases and found better solutions in 20% of cases.

Google DeepMind then applied AlphaEvolve to a handful of real-world problems. As well as coming up with a more efficient algorithm for managing computational resources across data centers, the tool found a way to reduce the power consumption of Google’s specialized tensor processing unit chips.

AlphaEvolve even found a way to speed up the training of Gemini itself, by producing a more efficient algorithm for managing a certain type of computation used in the training process.

Google DeepMind plans to continue exploring potential applications of its tool. One limitation is that AlphaEvolve can’t be used for problems with solutions that need to be scored by a person, such as lab experiments that are subject to interpretation.

Moosbauer also points out that while AlphaEvolve may produce impressive new results across a wide range of problems, it gives little theoretical insight into how it arrived at those solutions. That’s a drawback when it comes to advancing human understanding.

Even so, tools like AlphaEvolve are set to change the way researchers work. “I don’t think we are finished,” says Kohli. “There is much further that we can go in terms of how powerful this type of approach is.”

Stay Ahead

Explore More Insights

Stay ahead with more perspectives on cutting-edge power, infrastructure, energy, bitcoin and AI solutions. Explore these articles to uncover strategies and insights shaping the future of industries.

IBM sends signals with its $10 billion quantum pledge

“A $10 billion investment is pretty significant,” said IDC analyst Heather West. “And it’s sending signals out that in order to actually move the technology forward at a significant pace and get to these larger systems, there has to be a bigger investment in the technology itself. If the US

NetBox at 10: Network inventory tool now a full infrastructure intelligence platform

NetBox Data Exchange (NDX): A database of infrastructure component metadata covering device logical characteristics, lifecycle dates, environmental data, and observability profiles across tens of thousands of device types. NetBox Asset Lifecycle: A procurement pipeline connecting network design to physical deployment through bills of materials, purchase orders, shipment tracking, and spares

How Jeetu Patel made Cisco unrecognizable

From dashboard sprawl to Cloud Control The most visible proof point of the new Cisco is Cloud Control, the unified management plane that now spans networking, security, compute, observability, collaboration, and an expanding ecosystem of third-party tools. Cisco is careful to note that this is not just another single pane

IBM, ServiceNow team to bring AI to legacy enterprise systems

Decades of deeply interconnected legacy systems are the biggest barrier to moving fast on AI, the companies stated. Their pairings will take advantage of Big Blue’s expertise in working with large systems, such as its mainframe environment, and extensive legacy applications, along with ServiceNow’s workflow and agent management platforms. “Most

Energy Secretary Keeps Coal-Fired Power Generation Alive in the Northwest

WASHINGTON—U.S. Secretary of Energy Chris Wright today issued an emergency order to keep affordable, reliable, and secure coal generation online and address critical grid reliability issues facing the Northwestern region of the United States. The emergency order directs TransAlta Centralia Generation LLC (TransAlta) to ensure that Unit 2 of the Centralia Generating Station in Centralia, Washington, a coal-fired power plant, remains available to operate. Centralia Unit 2 was scheduled to shut down at the end of 2025. The order minimizes the risk and cost of unnecessary blackouts. “Taking reliable generation off the grid compromises energy reliability and needlessly raises energy costs for Americans,” said Energy Secretary Wright. “During peak summer demand, Northwesterners deserve continued access to affordable, reliable, and secure energy to power and cool their homes.” Thanks to President Trump’s leadership, coal plants across the country are being saved from premature retirement and reversing plans to shut down. In 2025, more than 17 gigawatts of coal-power electricity generation were saved from going offline. As outlined in DOE’s Resource Adequacy Report, power outages could increase by 100 times by 2030 if the U.S. continues to take reliable power offline. The availability of Centralia to operate will continue to be an asset to maintain reliability in the Western Electricity Coordinating Council (WECC) Northwest region. The North American Electric Reliability Corporation’s (NERC) 2025 Long-Term Reliability Assessment assessed that the WECC Northwest region is at high risk of energy shortfalls over the next five years, noting that “rapid forecasted demand growth is driving the need for more resources” and that “periods of unserved energy are projected for both summer and winter.” This order is in effect beginning on June 15, 2026, through September 12, 2026. Background: According to the U.S. Environmental Protection Agency’s data, in 2025, Centralia generated an average of approximately 340,000 MWh per month, providing vital generation capacity to the region.  ###

United States, Cyprus, Greece, Israel and Rice University To Establish Eastern Mediterranean Energy Center in Houston

HOUSTON, TEXAS—U.S. Secretary of Energy Chris Wright today signed a Declaration of Intent (DOI) with the Minister of Energy, Commerce, and Industry of the Republic of Cyprus Michael Damianos, Minister of Environment and Energy for Greece Stavros Papastavrou, Israeli Ambassador to the United States Dr. Yechiel Leiter, and President of Rice University Reginald DesRoches to establish the Eastern Mediterranean Energy Center (EMEC). The agreement establishes a framework to strengthen cooperation between the respective nations through the Eastern Mediterranean Energy Center (EMEC). It also advances a key initiative envisioned under Secretary Rubio’s Eastern Mediterranean Security and Energy Partnership Act of 2019. The agreement advances President Trump’s commitment to strengthening America’s partnerships with key allies while expanding opportunities for U.S. energy development, innovation, and investment. As global energy demand continues to grow, the United States, Cyprus, Greece, and Israel will work together to promote energy security, strengthen critical infrastructure, support emerging technologies, and advance long-term economic growth throughout the Eastern Mediterranean. “The Eastern Mediterranean Energy Center will help fulfill President Trump’s vision of prosperity and energy security at home and abroad,” said Secretary Wright. “The Eastern Mediterranean is an increasingly important region for global energy development, and this agreement strengthens cooperation among key allies while advancing our shared goals of energy abundance, economic prosperity, and regional stability. By establishing the Eastern Mediterranean Energy Center at Rice University in Houston, we are ensuring all member nations of this agreement will benefit from a lasting partnership bound together by the brightest minds and industry leaders in hydrocarbon development.” The partnership will support collaboration on shared priorities including natural gas development, U.S. LNG infrastructure, energy transportation networks, grid reliability, critical infrastructure resilience, and emerging technologies. It will also facilitate scientific and technical exchanges, research partnerships, workforce development initiatives, and engagement with industry stakeholders. The Trump

Energy Secretary Secures Carolinas’ Grid Ahead of Period of Hot Weather

WASHINGTON—The U.S. Department of Energy (DOE) today issued an emergency order to mitigate blackouts in the Carolinas’ ahead of a period of hot weather. Issued pursuant to Section 202(c) of the Federal Power Act, the order authorizes Duke Energy Carolinas, LLC (“DEC”) and Duke Energy Progress, LLC (“DEP”) (collectively, “Duke Energy”) to operate specified units located within Duke Energy’s service territory to operate up to their maximum generation output levels, notwithstanding air quality or other permit limitations arising under federal, state, or local law or regulation, or other applicable source of law. The order was issued subsequent to Duke Energy’s application. The order will mitigate the risk of unnecessary blackouts brought on by unusually high load forecasts and high temperatures across the region. “Maintaining affordable, reliable, and secure power in the Duke Energy service territory is non-negotiable,” said U.S. Secretary of Energy Chris Wright. “The previous administration’s energy subtraction policies weakened the grid, leaving Americans more vulnerable during events like this. Thanks to President Trump’s leadership, we are reversing those failures and using every available tool ensuring Americans in the Carolinas’ have continued access to affordable, reliable, and secure energy to power and cool their homes.” On day one, President Trump declared a national energy emergency after the Biden administration’s energy subtraction agenda left behind a grid increasingly vulnerable to blackouts. The order is in effect beginning at 4:00 PM ET on June 11, 2026, and shall expire at 10:00 PM ET on June 12, 2026. Background: Duke Energy stated that some generating units are limited in providing needed generation because of conditions and limitations in their environmental permits. As a result, the system “may not have sufficient generation available to meet this unusually high demand and [Duke Energy] may be forced to curtail load in order to maintain security

Energy Department Issues RFP to Advance President Trump’s 172-Million-Barrel Strategic Petroleum Reserve Exchange

WASHINGTON—The U.S. Department of Energy (DOE) today issued a Request for Proposal (RFP) for an exchange of up to 40 million barrels of crude oil from the Strategic Petroleum Reserve (SPR). Today’s solicitation opens competitive bidding, continuing DOE’s execution of President Trump’s 172-million-barrel release as part of a coordinated 400-million-barrel action by International Energy Agency (IEA) member nations’ strategic reserves. Under President Trump’s leadership, DOE has advanced an unprecedented series of large-scale SPR exchange solicitations at record speed. These actions have moved critical crude oil supplies into the market to address short term supply disruptions and bolster energy security for the United States and its allies. The crude oil will originate from the SPR’s Big Hill and Bryan Mound sites. This action builds on the Department’s four previous solicitations that collectively awarded more than 133 million barrels across three completed exchanges. DOE’s earlier exchanges demonstrated the SPR’s ability to rapidly deliver crude under emergency authorities while achieving a 26 percent premium in returned barrels—expanding the reserve at no additional cost to American taxpayers. “With today’s announcement, we are accelerating the President’s commitment to a coordinated and strategic release that stabilizes global oil markets,” said DOE Acting Assistant Secretary for the Hydrocarbons and Geothermal Energy Office Curt Coccodrilli. “This exchange will help move oil swiftly to refiners, ease short-term supply pressures, and ensure the Strategic Petroleum Reserve continues to grow stronger through the return of premium barrels.” Under DOE’s exchange authority, participating companies will return the 40 million borrowed barrels with additional premium barrels, ensuring immediate market supply while increasing the SPR’s long-term inventory. Bids for this solicitation are due no later than 11:00 A.M. Central Time on Monday, June 15, 2026. For more information on the SPR, please visit DOE’s website.

DOE’s Hydrocarbons and Geothermal Energy Office Invests $3.6 Million to Modernize America’s Coal-Fired Power Plants

WASHINGTON—The U.S. Department of Energy’s (DOE) Hydrocarbons and Geothermal Energy Office (HGEO) today announced $3.6 million for nine design and engineering projects that will support the refurbishment or retrofit of existing coal power plants with transformational technologies that address wastewater systems and improve the efficiency, reliability, flexibility, and performance of coal and natural gas use. By upgrading our nation’s existing coal facilities, these initiatives will help strengthen the backbone of America’s power grid and ensure all American’s have access to affordable, reliable, and secure energy when they need it most. These efforts help to advance President Trump’s Executive Orders Reinvigorating America’s Beautiful Clean Coal Industry and Strengthening the Reliability and Security of the United States Electric Grid to restore common-sense energy policies that prioritize dependable power, affordability, and American workers. “America’s coal fleet is an undeniable pillar of our energy dominance and economic strength, but for too long, policies have undermined this vital industry and the dedicated workforce behind it, threatening our grid’s stability and driving up costs for everyday Americans,” said DOE Acting Assistant Secretary of the Hydrocarbons and Geothermal Energy Office Curt Coccodrilli. “With the project investments announced today, we are decisively moving to champion our existing coal plants, ensuring they continue to deliver affordable, reliable power, keep the lights on, and fuel America’s progress for generations to come.” Projects have been selected under three topic areas to provide a path forward to rapidly and cost-effectively restore the stability of the nation’s bulk power system while also finding beneficial uses for wastes generated by coal-based energy production. The projects will be executed in three phases, with design and engineering completed in Phase I, final engineering and detailed design completed in Phase II, and technology implementation and validation completed in Phase III. Selectees to receive Phase I funding include: Baker Hughes Energy Transition LLC (Houston, Texas),

Energy Department Releases Finalized Fusion Science and Technology Roadmap to Accelerate Commercial Fusion Power

WASHINGTON—The U.S. Department of Energy (DOE) today released the finalized Fusion Science and Technology (FS&T) Roadmap, a national strategy to accelerate the development and commercialization of fusion energy on the most rapid, responsible timeline in history. Building on earlier roadmap efforts, the finalized roadmap brings together fusion science, technology, infrastructure, workforce development, and commercialization priorities into a single national strategy to support fusion pilot plants and commercial fusion power in the mid-2030s. Fusion is the process that powers the sun and stars. For decades, scientists and engineers have worked to bring that same process to Earth as a source of abundant, reliable energy. The finalized roadmap outlines how DOE, industry, universities, and national laboratories will work together to accelerate the path toward commercial fusion energy in the United States. This effort advances President Trump’s energy dominance agenda and reinforces the Administration’s commitment to expanding reliable American energy production, strengthening domestic supply chains, and maintaining U.S. leadership in critical technologies. By accelerating progress toward commercial fusion power, DOE is helping secure a future of abundant and reliable energy. “Fusion energy has entered a new era defined by extraordinary scientific progress and public-private momentum,” said DOE Under Secretary for Science Dr. Darío Gil. “With this roadmap, we now have the clarity, coordination, and sustained commitment needed to turn the promise of fusion into a reality for the American people.” Developed with input from more than 800 scientists and engineers across the public and private sectors, the finalized FS&T Roadmap reflects contributions from more than 15 private companies, over 10 National Laboratories, and more than 70 universities. The roadmap identifies the critical science and technology gaps that must be closed to realize fusion pilot plants and strengthen U.S. leadership in the global fusion industry. The FS&T Roadmap establishes a unified strategy for the U.S.

Amazon claims its data centers are 7x more water-efficient than the industry average

“Amazon is on the leading edge, but it’s not a secret recipe,” he said. What sets the company apart is scale, execution, facility design, geographic mix, and its aggressive pursuit of energy goals. Others are doing the similar things, if through different avenues: Microsoft is investing in closed-loop cooling systems that dramatically reduce evaporative water loss. Google is heavily focused on reclaimed water and using AI to optimize data centers. Meta has long relied on outside-air cooling. And overall, the industry is moving toward liquid cooling for dense AI deployments, “which changes the water equation again,” said Kimball. One of the big variables is location: Climate influences water efficiency, so where a company builds its infrastructure is as important as its cooling methods. Further, power-consumptive AI changes the discussion, he emphasized; traditional enterprise workloads and dense AI training clusters create very different thermal profiles.

Marvell announces 102.4 Tbps switch silicon built for AI

Data movement has become an important concern in modern AI data centers. In the past, a cluster of a few servers could adequately handle back-office applications and databases. But with AI’s gigantic models, all sections of the data center need to move and receive data at high speeds. That requires a lot more power use than in the past. GPU- and XPU-based systems are approaching 120KW per rack, and switching and networking components consume approximately 15-25% of total rack power, making low-power switch silicon a strategic requirement. The Teralynx T100 delivers up to 25% lower power consumption than competitive solutions at a higher data rate. This enables AI infrastructures to deploy more accelerators within existing power envelopes without requiring additional power infrastructure. “As AI workloads evolve and scale exponentially, hyperscalers require network architectures that optimize latency, power and scalability simultaneously,” said Rishi Chugh, vice president and general manager of the data center switch business unit at Marvell, in a statement.

From the data center to the edge: How to build secure, effective enterprise AI infrastructure

While hyperscalers and neo-cloud providers may get the lion’s share of attention for providing AI infrastructure, many enterprises are taking a build-it-themselves approach to meet their specific AI requirements. The success of such projects is crucial to achieving business objectives, yet companies face significant challenges as they try to scale pilots to production. Organizations must keep up with the dynamic, ever-changing demands that AI applications place on compute and network infrastructure, from the data center to the edge. That means architecting systems to grow as demand warrants and to avoid performance bottlenecks. The architecture must also account for AI-driven security vulnerabilities and ensure appropriate defenses are in place. Yes, it’s a tall order. But here, in simplified form, is a three-step plan for meeting those objectives. Step one: Go modular Integrating all the required components in piecemeal fashion for an AI factory is complex, costly, and fraught with integration risk. Start with a modular design, based on proven NVIDIA reference architectures. A modular approach combines pre-validated accelerated computing hardware, AI software, and orchestration platforms, as well as networking and storage capabilities. A modular strategy speeds implementation and creates a faster time to value for your AI infrastructure. Using modules that combine compute, networking, and storage makes it easier to scale capacity as needed, whether in the data center or at edge facilities. In addition, the modular approach simplifies the job of addressing varying requirements, from inferencing engines at the edge to massive-scale model training in the data center, while staying within the same solution family. The same applies to easing integration processes, as modular platforms offer pre-validated software. The Cisco Secure AI Factory with NVIDIA approach, for example, includes hardware (Cisco AI PODS) that is pre-validated to work with NVIDIA AI Enterprise software; Cisco Security and Splunk Observability software; orchestration

OpenAI weighs Nvidia-backed lease for 10 GW Ohio data center campus

OpenAI would control the computing equipment under a 20-year lease and begin payments once the site starts operating, with the first phase expected in 2028. Nvidia is expected to supply the hardware and guarantee both OpenAI’s lease obligations and the developer’s financing, the report added. The reported structure highlights a broader shift in AI infrastructure strategy, where model developers, chip suppliers, and energy providers are forging increasingly long-term partnerships to secure compute capacity amid surging demand. “These types of symbiotic deals are becoming the norm as AI infrastructure rolls out,” said Neil Shah, vice president for research and partner at Counterpoint Research. “If a CIO picks OpenAI to be the base layer, they shouldn’t just accept whatever infrastructure comes with it. CIOs need to negotiate and demand that OpenAI uses a mix of capacity so all your eggs are not in one premium basket like Nvidia.” OpenAI and Nvidia did not immediately respond to requests for comment.

Arista unveils 1.6T rack-scale switch family for AI infrastructure

The new Arista family joins a growing ecosystem of vendors looking to tap into the 1.6T Ethernet world, which includes Cisco, Nvidia, Celestica and others. “Arista Network’s new 7060XE7 Series is a strong signal of where large-scale AI fabrics are heading: higher bandwidth, better power efficiency, and tighter integration between compute, optics, silicon, cooling, and network operating software,” wrote Sameh Boujelbene, vice president, data center switch and AI networks market research for Dell Oro, in a LinkedIn post. Among the features that stand out to her are “strong customer and ecosystem validation from Microsoft Azure, Oracle Cloud Infrastructure, Meta, AMD, and Broadcom.”

Water Emerges as a Critical Constraint for AI Data Centers

“There really has been a major shift within the last couple of years,” Bajpayee said. “I would even say within the last 12 months is where we have seen suddenly a rapid increase in the data center operators’ desire to control their water destiny.” For Gradiant, the MIT-born water technology company that built its reputation serving semiconductor manufacturers, pharmaceutical companies, and industrial customers worldwide, that shift has translated into a rapidly expanding pipeline of data center opportunities. More importantly, Bajpayee believes it signals a fundamental change in how the industry thinks about water itself. The conversation is no longer centered primarily on sustainability metrics or corporate environmental goals. Instead, operators increasingly view water as a business continuity issue. “We’re seeing operators themselves come to us and tell us that these are issues they are facing,” Bajpayee said. “They want to make sure they don’t get stalled, their permits don’t get pulled, their business doesn’t get stopped, and communities don’t push them out because they didn’t figure out a way to control their water.” From Water Treatment to Water Strategy That shift is occurring as Gradiant expands deployments of its recently announced HyperSolved platform, an end-to-end cooling water management system purpose-built for AI data centers. The company says HyperSolved is now being deployed with several of the world’s largest hyperscale operators across North America, Europe, and Asia, reflecting growing industry demand for integrated approaches to water infrastructure. While compute, networking, and power systems have evolved rapidly during the AI era, water management often remains fragmented, requiring operators to coordinate multiple vendors responsible for sourcing, treatment, cooling, wastewater management, reuse, discharge, and regulatory compliance. Gradiant’s approach seeks to consolidate those functions into a single integrated platform and operating model. The timing reflects the growing scale of the challenge. New AI data center

Microsoft will invest $80B in AI data centers in fiscal 2025

And Microsoft isn’t the only one that is ramping up its investments into AI-enabled data centers. Rival cloud service providers are all investing in either upgrading or opening new data centers to capture a larger chunk of business from developers and users of large language models (LLMs). In a report published in October 2024, Bloomberg Intelligence estimated that demand for generative AI would push Microsoft, AWS, Google, Oracle, Meta, and Apple would between them devote $200 billion to capex in 2025, up from $110 billion in 2023. Microsoft is one of the biggest spenders, followed closely by Google and AWS, Bloomberg Intelligence said. Its estimate of Microsoft’s capital spending on AI, at $62.4 billion for calendar 2025, is lower than Smith’s claim that the company will invest $80 billion in the fiscal year to June 30, 2025. Both figures, though, are way higher than Microsoft’s 2020 capital expenditure of “just” $17.6 billion. The majority of the increased spending is tied to cloud services and the expansion of AI infrastructure needed to provide compute capacity for OpenAI workloads. Separately, last October Amazon CEO Andy Jassy said his company planned total capex spend of $75 billion in 2024 and even more in 2025, with much of it going to AWS, its cloud computing division.

John Deere unveils more autonomous farm machines to address skill labor shortage

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Self-driving tractors might be the path to self-driving cars. John Deere has revealed a new line of autonomous machines and tech across agriculture, construction and commercial landscaping. The Moline, Illinois-based John Deere has been in business for 187 years, yet it’s been a regular as a non-tech company showing off technology at the big tech trade show in Las Vegas and is back at CES 2025 with more autonomous tractors and other vehicles. This is not something we usually cover, but John Deere has a lot of data that is interesting in the big picture of tech. The message from the company is that there aren’t enough skilled farm laborers to do the work that its customers need. It’s been a challenge for most of the last two decades, said Jahmy Hindman, CTO at John Deere, in a briefing. Much of the tech will come this fall and after that. He noted that the average farmer in the U.S. is over 58 and works 12 to 18 hours a day to grow food for us. And he said the American Farm Bureau Federation estimates there are roughly 2.4 million farm jobs that need to be filled annually; and the agricultural work force continues to shrink. (This is my hint to the anti-immigration crowd). John Deere’s autonomous 9RX Tractor. Farmers can oversee it using an app. While each of these industries experiences their own set of challenges, a commonality across all is skilled labor availability. In construction, about 80% percent of contractors struggle to find skilled labor. And in commercial landscaping, 86% of landscaping business owners can’t find labor to fill open positions, he said. “They have to figure out how to do

2025 playbook for enterprise AI success, from agents to evals

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More 2025 is poised to be a pivotal year for enterprise AI. The past year has seen rapid innovation, and this year will see the same. This has made it more critical than ever to revisit your AI strategy to stay competitive and create value for your customers. From scaling AI agents to optimizing costs, here are the five critical areas enterprises should prioritize for their AI strategy this year. 1. Agents: the next generation of automation AI agents are no longer theoretical. In 2025, they’re indispensable tools for enterprises looking to streamline operations and enhance customer interactions. Unlike traditional software, agents powered by large language models (LLMs) can make nuanced decisions, navigate complex multi-step tasks, and integrate seamlessly with tools and APIs. At the start of 2024, agents were not ready for prime time, making frustrating mistakes like hallucinating URLs. They started getting better as frontier large language models themselves improved. “Let me put it this way,” said Sam Witteveen, cofounder of Red Dragon, a company that develops agents for companies, and that recently reviewed the 48 agents it built last year. “Interestingly, the ones that we built at the start of the year, a lot of those worked way better at the end of the year just because the models got better.” Witteveen shared this in the video podcast we filmed to discuss these five big trends in detail. Models are getting better and hallucinating less, and they’re also being trained to do agentic tasks. Another feature that the model providers are researching is a way to use the LLM as a judge, and as models get cheaper (something we’ll cover below), companies can use three or more models to

OpenAI’s red teaming innovations define new essentials for security leaders in the AI era

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More OpenAI has taken a more aggressive approach to red teaming than its AI competitors, demonstrating its security teams’ advanced capabilities in two areas: multi-step reinforcement and external red teaming. OpenAI recently released two papers that set a new competitive standard for improving the quality, reliability and safety of AI models in these two techniques and more. The first paper, “OpenAI’s Approach to External Red Teaming for AI Models and Systems,” reports that specialized teams outside the company have proven effective in uncovering vulnerabilities that might otherwise have made it into a released model because in-house testing techniques may have missed them. In the second paper, “Diverse and Effective Red Teaming with Auto-Generated Rewards and Multi-Step Reinforcement Learning,” OpenAI introduces an automated framework that relies on iterative reinforcement learning to generate a broad spectrum of novel, wide-ranging attacks. Going all-in on red teaming pays practical, competitive dividends It’s encouraging to see competitive intensity in red teaming growing among AI companies. When Anthropic released its AI red team guidelines in June of last year, it joined AI providers including Google, Microsoft, Nvidia, OpenAI, and even the U.S.’s National Institute of Standards and Technology (NIST), which all had released red teaming frameworks. Investing heavily in red teaming yields tangible benefits for security leaders in any organization. OpenAI’s paper on external red teaming provides a detailed analysis of how the company strives to create specialized external teams that include cybersecurity and subject matter experts. The goal is to see if knowledgeable external teams can defeat models’ security perimeters and find gaps in their security, biases and controls that prompt-based testing couldn’t find. What makes OpenAI’s recent papers noteworthy is how well they define using human-in-the-middle

Stay Ahead, Stay ONMINE