Stay Ahead, Stay ONMINE

Inside the AI-optimized data center: Why next-gen infrastructure is non-negotiable

How are AI data centers different from traditional data centers? AI data centers and traditional data centers can be physically similar, as they contain hardware, servers, networking equipment, and storage systems. The difference lies in their capabilities: Traditional data centers were built to support general computing tasks, while AI data centers are specifically designed for […]

How are AI data centers different from traditional data centers?

AI data centers and traditional data centers can be physically similar, as they contain hardware, servers, networking equipment, and storage systems.

The difference lies in their capabilities: Traditional data centers were built to support general computing tasks, while AI data centers are specifically designed for more sophisticated, time and resource-intensive workloads. Conventional data centers are simply not optimized for AI’s advanced tasks and necessary high-speed data transfer.

Here’s a closer look at their differences:

AI-optimized vs. traditional data centers

  • Traditional data centers: Handle everyday computing needs such as web browsing, cloud services, email and enterprise app hosting, data storage and retrieval, and a variety of other relatively low-resource tasks. They can also support simpler AI applications, such as chatbots, that do not require intensive processing power or speed.
  • AI data centers: Built to compute significant volumes of data and run complex algorithms, ML and AI tasks, including agentic AI workflows. They feature high-speed networking and low-latency interconnects for rapid scaling and data transfer to support AI apps and edge and internet of things (IoT) use cases.

Physical infrastructure

  • Traditional data centers: Typically composed of standard networking architectures such as CPUs suitable for handling networking, apps, and storage.
  • AI data centers: Feature more advanced graphics processing units (GPU) (popularized by chip manufacturer Nvidia), tensor processing units (TPUs) (developed by Google), and other specialized accelerators and equipment.

Storage and data management

  • Traditional data centers: Generally, store data in more static cloud storage systems, databases, data lakes, and data lakehouses.
  • AI data centers: Handle huge amounts of unstructured data including text, images, video, audio, and other files. They also incorporate high-performance tools including parallel file systems, multiple network servers, and NVMe solid state drives (SSDs).

Power consumption

  • Traditional data centers: Require robust cooling systems such as air-based or raised floors, free cooling using outside air and water, and evaporative cooling. Methods depend on factors such as IT equipment density and energy efficiency/sustainability goals.
  • AI data centers: GPUs, due to their high processing power, generate much more heat and require advanced techniques such as liquid cooling, direct-to-chip cooling, and immersion cooling.

Cost

  • Traditional data centers: Use standard hardware and computing components that do take up a big chunk of IT budgets. Costs can be reduced with optimized components, processes, cloud resources, and diligence around energy use.
  • AI data centers: Are often far more expensive due to high costs of GPUs, ultra-high-speed networking components, and specialized cooling requirements.

Ultimately, AI-optimized and traditional data centers have pivotal, yet distinct, roles in enterprise. A key difference is adaptability: The rapid evolution of AI requires advanced infrastructure with modular designs that can accommodate evolving chip architectures, power densities, and cooling methods.

Key components of AI-optimized data centers

AI-ready data centers have specific requirements when it comes to infrastructure. They must be able to perform high-performance computing (HPC) and process enormous datasets for training, inference, deployment, and ongoing operation of AI systems. This process is enabled by:

  • AI accelerators: These specialized chips span hundreds, or even thousands, of servers working in tandem.
  • Fast and reliable networking: Low latency and high-bandwidth connections between compute clusters and storage and data sources is a must. In some cases, bandwidth requirements can reach into the terabits per second (Tbps). Leading providers incorporate direct cloud connectivity, software-defined networking, and high-speed, redundant fiber connections to support performance. Technologies such as ethernet and InfiniBand, and optical interconnects can quickly transfer data between chips, servers, and storage.
  • GPUs: Popularized by Nvidia and originally designed for rendering graphics in video games GPUs are electronic circuits that perform many calculations simultaneously, what’s known as parallel processing. This involves fragmenting complex tasks into smaller pieces that can be solved concurrently across multiple processors. Parallel processing makes GPUs fast, efficient, and scalable, optimizing neural networks and deep learning applications and reducing training and inference times.
  • TPUs, NPUs and DPUs: AI-ready data centers increasingly incorporate more specialized accelerators specifically built for AI workloads. These include tensor processing Units (TPUs), neural processing units (NPUs), and data processing units (DPUs).
    • TPUs speed up tensor computations, or multi-dimensional data structures, so that AI models can process complex data and perform calculations. They are extremely efficient at handling large-scale operations fundamental to training and running AI, and their high throughput and low latency make them ideal for AI and deep learning.
    • NPUs mimic the neural pathways of the brain, allowing for processing of AI workloads in real time. They are optimized for parallel processing and offload AI tasks from CPUs and GPUs to optimize performance, reduce energy needs, and support faster AI workflows.
    • DPUs offload and speed up networking, storage, and security functions, freeing up CPUs and GPUs to focus on AI tasks. DPUs often handle data compression, storage management, and encryption to help improve efficiency, security, and performance.

Advanced data center cooling systems

AI workloads produce a significant amount of heat, forcing a re-think of facility design, particularly when it comes to cooling. Energy-efficient techniques and advanced cooling systems are a must for AI-optimized data centers.

Shape
Shape
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.

Shape

AMD launches AI-targeted PCIe cards for current servers

Instinct MI350P PCIe cards are available in air-cooled systems with up to eight accelerator cards, which makes them ideal for small, medium, and large AI models for inference and RAG pipelines. It has 144GB of high bandwidth memory 3e (HBM3E) running at up to 4TB/s. Performance is estimated at 2,299

Read More »

Energy Department to Invest $36 Million in Enhanced Oil Recovery Program at the University of North Dakota

WASHINGTON—The U.S. Department of Energy’s (DOE) Hydrocarbons and Geothermal Energy Office (HGEO) today announced the selection of a new project with $36 million in federal funding for the University of North Dakota’s Energy & Environmental Research Center to advance the commercial deployment of enhanced oil recovery (EOR) technologies in the Bakken shale formation. Through an integration of laboratory, modeling, artificial intelligence (AI), and field-based activities, the Bakken Enhanced Oil Recovery–Cracking the Code (Bakken EOR-CC) program will generate critical data and insights to enable efficient, large-scale implementation of carbon dioxide-based EOR. These efforts support President Trump’s commitment to American energy dominance by advancing technologies that increase domestic energy production and deliver affordable, reliable, and secure energy for the American people. “North Dakota has proven itself to be a leader in energy innovation, and the Bakken Enhanced Oil Recovery initiative builds on that legacy,” said DOE Assistant Secretary for the Hydrocarbons and Geothermal Energy Office Kyle Haustveit. “This program is essential to maximize the full potential of our valuable hydrocarbon resources in the Bakken. By ‘cracking the code,’ these integrated pilot projects will help establish a clear path for the broad commercial deployment of enhanced energy recovery across the nation. The Bakken formation holds the potential to unlock billions of barrels of oil—resources that can power energy independence for generations to come.” The Bakken, a key unconventional tight oil play in North Dakota, holds substantial potential for increased oil recovery. Currently, only about 10% of the oil in unconventional shale formations is typically recovered. The Bakken EOR-CC program is designed to evaluate EOR strategies, potentially unlocking billions of additional barrels of incremental oil and extending the life of the state’s coal-fired power plants by utilizing their captured carbon dioxide for EOR. The program is uniquely positioned to capitalize on knowledge generated through six EOR pilot projects,

Read More »

OTC speakers say Venezuela reopening hinges on stability, legal clarity

The conversation comes as the Trump administration continues easing sanctions and encouraging American operators to re-engage with Venezuela’s oil and gas sector. Signs are growing that major international energy companies are reassessing opportunities in the country. ExxonMobil and ConocoPhillips have recently dispatched technical teams to evaluate oilfield infrastructure and upstream prospects, while Gulf Coast refiners have already increased imports of Venezuelan heavy crude. Panelists said the central question facing US energy companies is no longer whether Venezuela will reopen, but whether the conditions, pace, and overall risk profile of that reopening are sufficient to support large-scale, long-term capital investment. Speakers noted that Venezuela’s appeal extends far beyond short-term political change. The country holds one of the world’s largest and most diverse hydrocarbon resource bases, including extra-heavy crude in the Orinoco Belt, conventional light and medium oil, and significant offshore natural gas resources. The opportunity lies not only in the size of the resource base, but also in the long-term development potential, the panelists said. However, years of underinvestment, deteriorating infrastructure, and labor losses mean rebuilding the sector will require significant technical expertise and sustained capital commitments. Oilfield service companies are expected to play an important role if activity accelerates, particularly in offshore gas, heavy oil upgrading, drilling services, and infrastructure rehabilitation. Recent reports indicate service providers have already begun reactivating rigs and equipment stored in Venezuela in anticipation of renewed activity. Speakers emphasized that investors are seeking stable policies and durable legal frameworks before committing capital at scale. Trust in Venezuela’s legal and regulatory system remains weak following years of expropriations and contract disputes. Companies must evaluate not only Venezuela’s domestic political outlook, but also the broader geopolitical dynamics involving the US and China, Borrego noted. China’s long-standing investments and influence in Venezuela’s energy sector were referenced as an important

Read More »

NNPC advances rehab, expansion plans for idled refineries

Nigeria National Petroleum Corp. Ltd. (NNPC) has entered a memorandum of understanding (MOU) with China-based Sanjiang Chemical Co. Ltd. and Xinganchen (Fuzhou) Industrial Park Operation and Management Co. Ltd. for collaboration via a potential technical equity partnership to support ongoing rehabilitation and expansion plans at two of its currently idled in-country refining complexes. Announced in early May, the MOU’s proposed framework covers unspecified remaining rehabilitation works at subsidiary Warri Refining & Petrochemical Co. Ltd.’s (WRPC) 125,000-b/sd refinery in Nigeria’s Delta State and Port Harcourt Refining Co. Ltd.’s (PHRC) 60,000-b/sd hydroskimming refinery at Alesa-Eleme near Port Harcourt in Rivers State, NNPC said. Alongside operating and maintenance activities to help the sites achieve best-in-class, sustainable performance, the MOU also outlines proposed expansions and upgrades at both refineries to enable production of cleaner, higher-valued products, according to the company. While NNPC did not clarify the nature of expansion and upgrading plans for either of the refining sections of the sites, the operator said the potential collaboration with Sanjiang Chemical and and Xinganchen (Fuzhou) Industrial Park also would weigh options for expanding the two complex’s petrochemical capabilities, as well as future development of co-located, gas-based industrial hubs at the two locations. NNPC said formal signing of the MOU follows more than 6 months of technical and management discussions with the two Chinese firms to develop a roadmap for restoring sustained, high-performance manufacturing operations at both sites. The MOU comes as part of NNPC’s broader mission to identify potential privately held technical equity partners to help support rehabilitation and expansion of its existing but nonoperational refining infrastructure, which ideally would include a willingness to evaluate opportunities for adding co-located petrochemical production and gas-based industries at the sites, the operator said. The agreement with Sanjiang Chemical and and Xinganchen (Fuzhou) Industrial Park follows NNPC’s announcement earlier

Read More »

Oil prices touch 4-year high as Iran crisis continues

Brent crude briefly surged above $126/bbl on Apr. 30, marking a 4-year high, as escalating tensions surrounding the Iran conflict intensified concerns over prolonged disruptions to Middle East oil flows. Stay updated on oil price volatility, shipping disruptions, LNG market analysis, and production output at OGJ’s Iran war content hub. The price spike reflects a market increasingly driven by geopolitical risk. The ongoing US–Iran standoff has effectively curtailed shipping activity through the Strait of Hormuz, a critical artery for global crude trade, with negotiations showing little progress toward reopening the route. President Donald Trump reiterated that a US naval blockade of Iran would remain in place until Tehran abandons its nuclear ambitions, reinforcing expectations of a prolonged supply disruption. Amid these disruptions, US crude exports have surged, highlighting the country’s growing role as a global swing supplier. According to the US Energy Information Administration (EIA), US crude exports rose to a record 6.44 million b/d in the latest reporting week, driving the US to become a net crude exporter on a weekly basis for the first time since World War II. The shift was accompanied by a 6.2 million-bbl draw in US commercial crude inventories, underscoring the tightness in global supply as buyers in Europe and Asia turn to US barrels to offset Middle East disruptions. Meanwhile, policy and macroeconomic signals added another layer of complexity. The Federal Reserve held interest rates steady in its Apr. 29 meeting, citing persistent uncertainty around inflation and growth, particularly as higher energy prices threaten to feed into broader economic conditions. US gasoline prices have followed crude higher, rising to $4.30/gal on Apr. 30, reflecting tightening refined product balances ahead of the summer driving season.

Read More »

Expand Energy prepared to slow completion work if prices weaken further

Expand produced nearly 7.44 bcfed during the first quarter (93% natural gas), which was up nearly 10% from the first three months of 2025. Of that production, 46% came from assets in the Haynesville basin (up from about 39% a year earlier), 37% from Northeast Appalachia and the remainder from Southwest Appalachia. Wichterich, who in February replaced Nick Dell’Osso while Expand’s directors search for a new permanent leader, is looking to have full-year 2026 production average about 7.5 bcfed while deploying between 11 and 12 rigs and six to seven completion crews. The current quarter will feature some seasonal curtailments, executives said, and production is expected to remain flat from early this year. “We are in the right place at the right time,” Wichterich said on the conference call. “Our assets are reaching 90% of the expected demand growth in this country and our Haynesville [operation] is sitting at the epicenter of growth because of the LNG market. We think we are in the best position to take advantage of that.” On the LNG front, Expand executives this week signed a 20-year sales and purchase agreement with Delfin FLNG Vessel 1 that calls for Expand to supply about 1.15 million tpa. That contract replaces previous deals with Delfin and Gunvor Group and calls for the gas to be sold at a Henry Hub price and to start flowing in 2031. Expand produced a first-quarter net profit of $1.16 billion on total revenues of $4.4 billion. Those numbers were an improvement from early 2025, when the company lost $249 million on $2.2 billion in sales, the latter figure hampered by $1 billion loss on derivatives. Shares of Expand (Ticker: EXE) were up nearly 3% to about $99.50 in afternoon trading on April 29. Over the last six months, they’re essentially flat

Read More »

Woodside appoints Lonnie as EVP, COO Australia

@import url(‘https://fonts.googleapis.com/css2?family=Inter:[email protected]&display=swap’); .ebm-page__main h1, .ebm-page__main h2, .ebm-page__main h3, .ebm-page__main h4, .ebm-page__main h5, .ebm-page__main h6 { font-family: Inter; } body { line-height: 150%; letter-spacing: 0.025em; } button, .ebm-button-wrapper { font-family: Inter; } .label-style { text-transform: uppercase; color: var(–color-grey); font-weight: 600; font-size: 0.75rem; } .caption-style { font-size: 0.75rem; opacity: .6; } #onetrust-pc-sdk [id*=btn-handler], #onetrust-pc-sdk [class*=btn-handler] { background-color: #c19a06 !important; border-color: #c19a06 !important; } #onetrust-policy a, #onetrust-pc-sdk a, #ot-pc-content a { color: #c19a06 !important; } #onetrust-consent-sdk #onetrust-pc-sdk .ot-active-menu { border-color: #c19a06 !important; } #onetrust-consent-sdk #onetrust-accept-btn-handler, #onetrust-banner-sdk #onetrust-reject-all-handler, #onetrust-consent-sdk #onetrust-pc-btn-handler.cookie-setting-link { background-color: #c19a06 !important; border-color: #c19a06 !important; } #onetrust-consent-sdk .onetrust-pc-btn-handler { color: #c19a06 !important; border-color: #c19a06 !important; } Woodside Energy has appointed Breyden Lonnie as executive vice-president and chief operating officer Australia. Lonnie has acted in the role since December 2025, with responsibility for Woodside’s portfolio of operations and projects in Western Australia and the Bass Strait. He joined the company in 2005. Prior, Lonnie—who has 24 years of industry experience in engineering, operations, planning, and management roles in Australia and internationally—served as Woodside’s vice-president North West Shelf. He joined Woodside in 2005. In a post to his LinkedIn account, Lonnie said “2026 will be an important year for [Woodside’s] Australian business, with a strong focus on disciplined operations, progressing Scarborough toward first LNG and taking operatorship of the Gippsland Basin Joint Venture.” 

Read More »

AWS hit by US-East-1 outage after data center thermal event

AWS shifted traffic away from the affected zone for most services and warned of longer-than-usual provisioning times. As the evening progressed, the company struggled to bring temperatures down. By 6:47 PM PDT, AWS warned that “Other AWS services that depend on the affected EC2 instances and EBS volumes in this Availability Zone may also experience impairments,” and at 8:06 PM PDT, it conceded that “progress is slower than originally anticipated,” recommending that customers needing immediate recovery restore from EBS snapshots or launch resources in unaffected zones. By 10:11 PM PDT, AWS reported “incremental progress to restore cooling systems” but said users were still “experiencing elevated error rates and latencies for some workflows.” The May 7 incident is not the first time US-EAST-1 has gone down. The region suffered two outages in October 2025, including a 15-hour disruption on October 19 and 20 caused by a race condition in DynamoDB’s automated DNS management system that affected over 70 AWS services and produced cascading failures across Slack, Atlassian, Snapchat, and other dependent services. AWS regions in Ohio have also experienced power-related outages tied to EC2 instances in past years.

Read More »

Lumen advances cloud networking vision with $475M Alkira buy

Lumen puts its total addressable market at approximately $70 billion once Alkira’s international and cloud-to-cloud coverage is included. “Alkira is a bull’s eye in terms of strategic alignment and value creation,” Johnson said. “For Lumen, we expect it to dramatically accelerate our road map execution from years to months.” How the architecture works Alkira operates as a cloud-native, carrier-agnostic control plane. Rather than relying on physical hardware at each interconnection point, it uses a virtual port model that lets enterprises design, deploy and manage network connectivity across clouds, data centers and on-premises environments through a single interface. Alkira is distinct from Lumen’s existing Project Berkeley, which introduces fabric ports for building-to-cloud on-ramp connectivity. “Fabric ports is about enabling building on-prem to be able to connect to the cloud and to be able to grow those services in a cloud economic way,” Johnson said. “The Alkira platform really focuses on the East-West interconnect. So that’s data center-to-data center, cloud-to-cloud, so they operate with more of a virtual port kind of a model, and it’s better together.” Lumen’s Multi-Cloud Gateway bridges the two domains, enabling customers to connect any cloud and any data center over Lumen’s private network. After close, Multi-Cloud Gateway and Alkira together are intended to give customers a single control plane for routing, policy and security across both north-south and east-west connectivity.

Read More »

Data Center Jobs: Engineering, Construction, Commissioning, Sales, Field Service and Facility Tech Jobs Available in Major Data Center Hotspots

Each month Data Center Frontier, in partnership with Pkaza, posts some of the hottest data center career opportunities in the market. Here’s a look at some of the latest data center jobs posted on the Data Center Frontier jobs board, powered by Pkaza Critical Facilities Recruiting. Looking for Data Center Candidates? Check out Pkaza’s Active Candidate / Featured Candidate Hotlist Power Applications Engineer Pittsburgh, PA This position is also available in: Denver, CO; Andrews, SC and remotely. Our client is a leading provider and manufacturer of industrial electrical power equipment used in industrial applications for mission critical operations. They help their customers save money by reducing energy and operating costs and provide solutions for modernizing their customer’s existing electrical infrastructure. This company provides cooling solutions to many of the world’s largest organizations and government facilities and enterprise clients, colocation providers and hyperscale companies. This career-growth minded opportunity offers exciting projects with leading-edge technology and innovation as well as competitive salaries and benefits. Electrical Commissioning Engineer New Albany, OH This traveling position is also available in: New York, NY; White Plains, NY;  Dallas, TX; Richmond, VA; Ashburn, VA; Montvale, NJ; Charlotte, NC; Atlanta, GA; Hampton, GA; Cedar Rapids, IA; Phoenix, AZ; Salt Lake City, UT; Kansas City, MO; Omaha, NE; Chesterton, IN or Chicago, IL. *** ALSO looking for a LEAD EE and ME CxA Agents and CxA PMs. ***  Our client is an engineering design and commissioning company that has a national footprint and specializes in MEP critical facilities design. They provide design, commissioning, consulting and management expertise in the critical facilities space. They have a mindset to provide reliability, energy efficiency, sustainable design and LEED expertise when providing these consulting services for enterprise, colocation and hyperscale companies. This career-growth minded opportunity offers exciting projects with leading-edge technology and innovation as well as

Read More »

Switch storm coming: Gartner forecasts price hikes, long lead times for enterprise data center switches

“If you’re a vendor and you’re doing what you’re supposed to do, you want to capture the growth,” he says. Zeus Kerravala, founder and principal analyst with ZK Research, agrees. “Cisco, Arista, Juniper and those companies that build data center equipment, make no mistake, their resources are directed towards AI first because they want to be part of those big buildouts,” he says. “There’s a lot of money being poured into neoclouds, things like that. They’ve reprioritized the resources based on where market demand is.” Price hikes, long lead times, sketchy support The repercussions for companies with traditional data centers include higher prices, long lead times, and perhaps subpar support. Gartner predicts switch price increases of 15% to 40%, largely the result of resource constraints, and lead times of three to nine months, up from one to two months in mid-2025. Constraints should ease by around the middle of next year, but don’t expect prices to come down. “Generally speaking, vendors have no consistent track record of reducing prices in these networking markets,” Lerner says. At the same time, with vendors dedicating scarce engineering talent to AI, they likely won’t invest in significant innovations for non-AI switch families. The same goes for support.

Read More »

Build Fast, Pay Your Way: Washington’s AI Infrastructure Doctrine

In the first quarter of 2026, the U.S. government made one point unmistakable. Washington wants more data center capacity, more AI infrastructure, and more domestic power. But it no longer views these projects as conventional commercial real estate. Across the White House, DOE, FERC, EPA, EIA, and the federal permitting apparatus, data centers are now being treated as strategic infrastructure. That designation brings tangible support in the form of faster permitting, access to federal land, and a more explicit embrace of large-scale power development. It also comes with conditions: stricter expectations around who funds transmission upgrades, who provides new generation, how water is managed, and how much operational data operators must disclose. This is the new federal posture: accelerate the buildout, but impose discipline on its consequences. Washington is not pulling back in the face of local opposition. It is pushing forward, while making clear that the next phase of data center growth must carry its own infrastructure burden. Who Will Pay? The question is no longer whether the United States will support the next wave of hyperscale and AI campus construction. The question is under what terms, and whether utilities, communities, and ratepayers will be asked to subsidize it. The outcome of that debate will be set less by local politics than by the federal rules now taking shape. The clearest signal came on March 4, when President Trump announced the “Ratepayer Protection Pledge.” Amazon, Google, Meta, Microsoft, OpenAI, Oracle, and xAI committed to “build, bring, or buy” new generation for their data centers and to fund the full cost of required grid and transmission upgrades. The administration also said those companies would coordinate with grid operators to provide backup generation in emergencies. The message was direct: data centers can grow, but the costs and reliability risks tied to

Read More »

300 MW Hyperscaler Lease Validates Applied Digital’s AI Infrastructure Financing Model

The Model Behind the Lease Applied Digital is packaging a full development solution for AI infrastructure: site, utility access, power distribution, cooling systems, and a financing framework capable of supporting multi-hundred-megawatt deployments. The approach reduces the integration burden on hyperscale customers and aligns delivery with the scale and timelines of AI demand. The Delta Forge 1 lease indicates that at least one major hyperscaler is willing to commit to that model on a long-term basis. The scale of the agreement reinforces that point. The lease accounts for 300 MW within a 430 MW campus, with capacity structured across two 150 MW buildings. The agreement spans two leases and includes three five-year renewal options, establishing a long-duration footprint at the site. This level of commitment effectively anchors the first phase of Delta Forge 1 and provides a clear validation of the campus’s initial buildout. Financing Follows the Lease Applied Digital paired the Delta Forge 1 tenant announcement with a financing update that underscores the link between signed demand and capital formation. The company expects to secure up to $600 million in additional funding, including a senior secured bridge facility of up to $300 million to support continued development at Polaris Forge 1, along with a $300 million revolving credit facility for development, working capital, and transaction expenses. The structure highlights how hyperscaler commitments can be translated into financing capacity across a broader platform. The Delta Forge 1 lease functions as a catalyst for the next phase of capital deployment. That momentum builds on a financing-heavy stretch. In its April 8 fiscal third-quarter results, Applied Digital disclosed a $2.15 billion private offering of 6.750% senior secured notes due 2031 to support Polaris Forge 2. The company also detailed credit enhancements tied to CoreWeave leases at Polaris Forge 1 following an investment-grade A3

Read More »

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.

Read More »

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

Read More »

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

Read More »

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

Read More »