Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

The US government is getting good at squeezing technology vendors and foundational model players OpenAI and Anthropic are raising to give away ChatGPT and Claude to the Feds for $1 a year.

It’s quite possible that the race to near-zero for LLM pricing isn’t going to stop with the General Services Administration (GSA) as scale enterprises follow suit.

Anthropic said it will offer Claude for Enterprise and Claude for Government to all branches of the US government for $1 per year. The deal is available for a year. OpenAI’s offer is OpenAI for Government for $1 per agency for the next year.

The deals make sense from a land-expand-perspective, but I can already write the headline when the $1 offers expire. The headlines will go like this: “OpenAI and Anthropic gouge US”. Now maybe none of these deals matter since OpenAI and Anthropic are already in a nice Department of Defense contract.

• Frontier AI companies land Department of Defense deals

Headline risk and the race to scale adoption at any cost are the two primary reasons OpenAI and Anthropic are going the value-meal route for LLMs. That headline risk is also giving GSA leverage. In 2025, the GSA announced agreements with AWS, DocuSign, Oracle, Elastic, Salesforce, Adobe, Google and Microsoft. According to the Congressional Budget Office, the US government employees 2.25 million military personnel and 2.25 million civilian workers.

Now what is the risk? Large enterprises may want better pricing too. We already know that LLMs are reaching the good enough stage.

• The rise of good enough LLMs

If I’m Walmart, the deals OpenAI and Anthropic gave the US government would have me prepping the “lower the price or else” emails. Walmart has the most workers at 2.1 million followed by Amazon at 1.56 million, according to CompaniesMarketCap. We assume Amazon will optimize LLM costs with Amazon Bedrock and its own Nova models. Walmart could squeeze its LLM providers (and tech vendors in general) just as it does suppliers.

We’ll assume that BYD and Foxconn, two companies with an employee base approaching 1 million, will go with Alibaba’s Qwen or DeepSeek.

Add it up and it’s not a reach to see LLM price pressure accelerating. With value-LLM deals, open-source models and private label models like Amazon Nova commoditization is just going to accelerate.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Every day, Constellation Research analysts are swapping takes, headlines, observations and broad themes on our internal watercooler thread. It’s part entertainment and part college class. Here’s a look at Holger Mueller and Esteban Kolsky’s recent debate on data, infrastructure and strategy.

Core Disagreement

• Mueller’s Position:
  • The foundation of enterprise success is data — "it starts with the data."
  • Data is continuously created (e.g., web searches) and must be the priority before infrastructure or models.
  • Correct and early adoption of technology can be a key differentiator, enabling growth in efficiency and effectiveness.
  • AI changes the game by making it possible to process data at a scale that was previously impossible.
• Kolsky’s Position:
  • Data matters, but it is not the most important focus in 2025.
  • After decades of investment in data warehouses, CDPs, and models, many enterprises still cannot move from "A to H" because they lack the right infrastructure and business models.
  • Technology alone is no longer a differentiator; the winners will be those who build forward-looking ecosystems and innovative business models on robust infrastructure.
  • Chasing every "new tech" is a resource trap — sustainable success requires a broader, multidisciplinary strategy.

Key Points

1. Data vs. Infrastructure
   • Mueller: Data is the car; without it, you can’t drive anywhere.
   • Kolsky: Infrastructure is the freeway; without it, even the fastest car (best data) is useless.
2. Tech as Differentiator
   • Mueller: Early, correct adoption of tech is a differentiator, and with AI accelerating tech’s capabilities, it is more important than ever.
   • Kolsky: Early adoption provides a temporary edge but leads to technical debt; tech is an enabler, not the differentiator.
3. Enterprise Strategy
   • Kolsky: Enterprises need to think beyond narrow tech or data focus — similar to how industry analysts must avoid siloed perspectives.
   • Mueller: Agrees on interconnectedness but emphasizes continuous cycles of evaluate → pilot → adopt → scale.
4. Sustainability of Early Adoption
   • Kolsky: Constantly “skating on the edge” isn’t sustainable.
   • Mueller: The key is avoiding the “efficiency trap” and practicing enterprise acceleration.

Common Ground

• Both agree technology is an enabler and everything is interconnected (data, infrastructure, business models).
• Both recognize AI’s impact on data processing and enterprise capabilities.
• Both stress the need for continuous improvement — but differ on whether the starting point should be data or infrastructure.

Recap

• Mueller: Data = car → you need the car first to travel.
• Kolsky: Infrastructure = freeway → without roads, the best car is useless.
• Disagreement lies in which comes first and where the main strategic focus should be.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Workday said it has acquired Flowise, a platform to create AI agents and automate workflows. Terms of the deal weren't disclosed. 

In a nutshell, Flowise has a low-code platform with modular building blocks ranging from workflows to autonomous agents. The Y Combinator backed company has a free tier, a starter version that runs $35 a month and a pro version for mid-sized companies for $65 a month. There's also an enterprise version.

Flowise appears to have developer cred with high ratings on GitHub and an affiliate program for referrals. Flowise's platform covers AI development from prototyping to building to evaluation and analytics in a flow to move to production.

For Workday, Flowise could give it the ability to more easily turn its data, processes and financial and human capital workflows into AI-powered autonomous agents.

Holger Mueller, an analyst at Constellation Research, said Flowise fills a gap at Workday.

"Workday needs to accelerate its AI roadmap and the acquisition of Flowise brings a key capability to Workday's AI plans. Flowise's strong workflow capabilities should allow Workday to convert many existing workflows into AI agents - a good start for helping customers. Now all eyes are on Workday Rising where the new AI strategy, roadmap and deliverables are likely to be presented."

• Workday delivers strong Q1, says 60% of customers using its AI
• Workday aims to be system of record for AI agents, digital labor |
• Workday Enters the Era of AI
• Employee Experience Platforms: Workday Peakon

Workday CTO Peter Bailis said Flowise's open-source foundation will give customers the ability to build, customize and deploy their own AI agents on the Workday platform. Given Workday's reach, the company should be able to scale Flowise, which operates in multiple verticals.

• Enterprise technology customers look to AI, efficiency to combat uncertainty
• What we learned from enterprise tech buyers in first half of 2025
• Lessons from early AI agent efforts so far
• Every vendor wants to be your AI agent orchestrator: Here's how you pick
• Agentic AI: Everything that’s still missing to scale

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Oracle Cloud Infrastructure customers will get access to Google Cloud's Gemini models in a new partnership between the companies.

Google Cloud and Oracle said they have expanded a partnership to give customers access to the latest Gemini models, including Gemini 2.5, via the OCI Generative AI service. Oracle added that it will made a wide range of Google Cloud models available within its Fusion cloud applications.

Oracle has been offering more models including xAI's Grok. Oracle said it will provide a new integration with Google Cloud's Vertex AI to tap into the Gemini models as well as industry models such as MedLM.

In a post, Oracle said the initial model rollout will include Gemini Ultra, Gemini Pro and Gemini Nano.

According to Oracle, OCI customers can use existing Oracle Universal Credits to apply to the Gemini models. OCI customers currently have access to models from Cohere, Meta Llama and xAI.

Oracle's cloud, AI plans are a master class in co-opetition

Google Cloud and Oracle forged a partnership last year to offer Oracle Database@Google Cloud along with deals with Microsoft Azure and AWS. That Oracle Database@Google Cloud has scaled into 2025.

Holger Mueller, an analyst at Constellation Research, said:

"This is an important triple win. It's a huge win for OCI customers as they get access to the leading multimodal LLM with Gemini, a win for Oracle as they can show they are becoming the 'Switzerland' for LLMs and for Google as it gets more mileage for Gemini. Key questions remain as where the training can happen, where the data will reside and more but this is a big win for OCI customers."

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Cohere raised $500 million in venture funding for a $6.8 billion valuation and hired a CFO and Chief AI Officer.

The funding lands as Cohere North, an agentic AI platform, scales up. The funding round was led by Radical Ventures and Inovia Capital with participation from existing investors such as AMD Ventures, Nvidia, PSP Investments and Salesforce Ventures.

Cohere added that Healthcare of Ontario Pension Plan was a new investor.

• Cohere North generally available
• Cohere launches North, aims to be an AI agent workspace
• Cohere rolls out Embed 4, an enterprise multimodal search model

As for the new Cohere hires, the company said Joelle Pineau, former head of Meta's Fundamental AI Research, will become Chief AI Officer. Francois Chadwick will become CFO. Chadwick is an alum of KPMG, Uber and Shield AI.

Along with Anthropic, Cohere is one of the leading large language model players focused on the enterprise. "We are at a pivotal moment in accelerating the delivery of secure AI that empowers enterprises worldwide, and we’re excited to enter this new phase of expansion alongside our partners," said Cohere CEO Aidan Gomez.

Cohere said the funding will be used to accelerate agentic AI use cases in businesses and governments primarily through North. The company added that strategic partners such as Oracle, Dell, RBC, Bell, Fujitsu, LG CNS and SAP are using North as a platform.

North uses Cohere's Command A, Command A Vision, Embed 4 and Rerank 3.5 models.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Five9 said it has integrated its customer experience and contact-as-a-service platform with Epic, one of the top electronic health record providers. The companies will launch Five9 Fusion for Epic in a bid to land more healthcare enterprises.

With the integration, Five9 will be included into Epic's Toolbox. The API integration between Five9 and Epic should enable agents to connect with patients on one screen. Epic has about 42% of the EHR market.

Constellation Research analyst Liz Miller noted that Five9's approach with Epic works well because it combines capabilities for an offering to healthcare. "Five9 has done the hard work around other aspects required of healthcare including security, identity assurance and other requirements," said Miller.

• ServiceNow, Salesforce invest $1.5 billion in Genesys as Five9 CEO retires
• Five9 bets contextual CX data will give it an AI edge

Fusion9 Fusion for Epic, which will be generally available in September, includes the following:

• Patient context for every call includes relevant data as soon as a call is received.
• One-click outbound calling to enable agents to initiate calls from Epic modules.
• HIPAA-compliant call recording controls.

The news lands after Five9 announced second quarter earnings and the retirement of CEO Mike Burkland after a successor is named. Five9 reported second quarter net income of $1.2 million, or a penny a share, on revenue of $283.3 million, up 12% from a year ago.

For 2025, Five9 is projecting revenue between $1.1435 billion and $1.1495 billion with non-GAAP earnings of $2.86 a share to $2.90 a share.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Cisco said it saw strong orders for AI infrastructure as the company reported a strong fourth quarter.

The company reported fourth quarter earnings of $2.8 billion, or 71 cents a share, on revenue of $14.7 billion, up 8% from a year ago. Non-GAAP earnings in the fourth quarter were 99 cents a share.

Wall Street was looking for fourth quarter earnings of 98 cents a share on revenue of $14.62 billion.

Chuck Robbins, CEO of Cisco, said "the AI infrastructure orders we received from webscale customers in fiscal 2025 were more than double our original target." He added that the company expects to see strong demand form sovereign AI deployments and emerging AI cloud providers. Robbins added that Cisco’s sales scale is boosting its Splunk unit and security strategy.

• Cisco tunes network portfolio, gear for AI agents, introduces AgenticOps
• Cisco launches quantum lab, unveils quantum networking chip

Cisco rival Arista surged last week after it said revenue growth will be about 25% in 2025 to $8.75 billion due to strong enterprise and AI demand. Jayshree V. Ullal, CEO of Arista, said during the company's second quarter earnings call that "we recognize the potential to build a truly transformational networking company." Juniper, another Cisco rival, has been acquired by HPE.

Here's a look at Cisco by the numbers:

• Fourth quarter orders were up 7% from a year ago, across multiple regions. AI infrastructure orders from webscale customers topped $800 million in the fourth quarter for a fiscal 2025 total of more than $2 billion.
• Cisco's networking business delivered revenue growth of 12% to $7.63 billion in the fourth quarter.
• Security revenue was up 9% in the fourth quarter to $1.95 billion.
• For fiscal 2025, Cisco reported earnings of $2.61 a share on revenue of $56.7 billion.

As for the outlook, Cisco projected first quarter non-GAAP earnings of 97 cents a share to 99 cents a share on revenue of $14.65 billion to $14.85 billion. Fiscal 2026 revenue will be between $59 billion to %60 billion with non-GAAP earnings of $4 a share to $4.06 a share.

Key items from Robbins during the conference call:

• “Our new smart switches are powered by Silicon One and deliver enhanced performance, quantum secure networking and radically simplified cloud native and AI driven operations, all supporting the new realities as AI changes how it work and collaborate. The introduction of our new switches marks the beginning of a major multiyear refresh cycle opportunity for Cisco's large installed campus switching base.”

• “AI orders are ramping and we have a growing pipeline in the hundreds of millions of dollars as these customers look to Cisco to provide simple, scalable and secure solutions.”
• “The Cisco secure AI factory with Nvidia provides a trusted blueprint for building secure AI ready data centers for enterprises, sovereign cloud providers and newly emerging neo cloud providers.”
• “While we have some clarity on tariffs, we are still operating in a complex component environment. Our Q1 and fiscal year 2026 guide assumes current tariffs and exemptions remain in place for the end of fiscal 2026.”
• “We're pretty confident that we haven't seen any indication of any pull forwards in AI infrastructure among backend networks of the cloud providers.”

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Rigetti Computing said its 36-qubit multi-chip quantum computer is generally available and outlined a plan to build a 100-qubit system with 99.5% fidelity by the end of 2025, but quantum advantage is likely 4 years away.

The technology milestones and outlook were outlined as Rigetti reported second quarter results that missed estimates.

Rigetti reported a second quarter net loss of $39.7 million, or 13 cents a share, on revenue of $1.8 billion, down from $3.09 million a year ago. Wall Street was expecting Rigetti to report a second quarter loss of 4 cents a share on revenue of $1.87 million

The company said second quarter revenue was impacted by the expiration of the national quantum initiative and pending reauthorization in the US Congress.

Quantum computing is an immature industry that generates a lot of headlines but many pure plays are just building revenue streams. Contracts--often with government agencies--are lumpy and so are the quarterly results. Given the boom in quantum stocks, many of the quantum computing vendors have been able to fortify balance sheets and raise cash.

For instance, Rigetti ended the second quarter with $576.1 million in cash and equivalents. IonQ also fortified its balance sheet with a $1 billion stock offering in July. IonQ, which has been on an acquisition spree, reported second quarter earnings last week with a net loss of $177.5 million on revenue of $20.7 million. IonQ had $1.6 billion of pro forma cash and equivalents as of July 9.

Speaking on a conference call with analysts, Rigetti CEO Dr. Subodh Kulkarni said the company's chiplet approach is scaling.

The company's Cepheus-1-36Q system is generally available on the Rigetti Quantum Cloud Services Platform and coming to Microsoft Azure. The system has reduced error rates from the Ankaa-3 system that preceded it and hits a median two-qubit gate fidelity of 99.5%.

"Just 6 months after our record performance with Ankaa-3, we've once again halved our error rates with Cepheus-1-36Q. We believe quadrupling our chiplet count and significantly decreasing error rates is the clear path towards quantum advantage and fault tolerance. We intend to continue this momentum with our 100 plus qubit system planned for the end of the year," said Kulkarni.

• IBM outlines quantum computing roadmap through 2029, fault-tolerant systems

Rigetti is betting on superconducting qubits as the ultimate leading technology for quantum computing. Other vendors have bet on competing technologies.

Kulkarni has been largely hype free when it comes to quantum computing. He noted that there are multiple challenges before quantum advantage is reached including dilution refrigerators and cabling as well as error correction.

He said:

"At this point, we believe we are still about 3 to 4 years away from getting to the 1,000-plus qubit, 99.9% fidelity with error correction and gate speeds of less than 50 nanoseconds, which is when we achieve quantum advantage. If we can accelerate that time line using our strengthened balance sheet we will obviously look at that. We are still looking at roughly about 4 years to get to that quantum advantage point."

• Constellation ShortListâ„¢ Quantum Computing Platforms | Quantum Computing Software Platforms | Quantum Full Stack Players
• CxOs need to focus on quantum computing readiness, not the noise

Key points from the earnings call from Kulkarni:

• "It's our view that superconducting qubits are the leading modality for quantum computers due to their ability to scale and their ability to achieve gate speeds more than 1,000x faster than other modalities like ion traps and pure atoms."

• "Transitioning from a monolithic chip to chiplets enables greater control over chip uniformity, which in turn improves performance. Leveraging chiplets also reduces manufacturing complexity and improves fabrication yield."
• "While we are pleased with our sequential growth in quarterly revenues, we believe achievement of our technology milestones remains a key metric to achieving our long-term success."
• "Our focus continues to be on R&D development. We will obviously look at every opportunity to accelerate our time line. Right now, we believe we are funding R&D adequately to hit the milestones that we have laid out. Our plan for the end of the year is to deliver a multichip 100- plus qubit system with 99.5% 2-qubit gate fidelity. And from there on to continue to increase the fidelity as well as qubit count using chiplet approach. Every opportunity we get to accelerate that time line, we continue to look at it and we will do so."

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

CoreWeave is building its AI infrastructure so it can easily switch back and forth from training to inference workloads as it works through strong demand and supply constraints.

The AI infrastructure provider reported a second quarter net loss of $290.51 million, or 60 cents a share, on $1.21 billion, up 207% from a year. Wall Street was looking for a loss of 49 cents a share on revenue of $1.08 billion.

The company exited the second quarter with a revenue backlog of $30.1 billion, double the backlog from a year ago. CoreWeave landed a $4 billion expansion deal with OpenAI in addition to a previously announced $11.9 billion contract.

• CoreWeave buys Core Scientific for $9 billion
• CoreWeave's IPO: What you need to know
• CoreWeave to buy AI developer Weights & Biases for a reported $1.4B

CoreWeave raised its 2025 revenue guidance to $5.15 billion to $5.35 billion. "We ended the quarter with nearly 470 megawatts of active power, and we increase the total contracting power approximately 600 megawatts to 2.2 gigawatts. We are aggressively expanding our footprint on the back of intensifying demand signals from our customers," said CoreWeave CEO Michael Intrator.

Here's a look at the takeaways.

Overall demand. CoreWeave signed expansion contracts with both of its hyperscale customer in the past eight weeks. "Our pipeline remains robust, growing and increasingly diverse, driven by a full range of customers, from media and entertainment to healthcare to finance to industrials and everything in between," said Intrator. "The proliferation of AI capabilities into new use cases and industries is driving increased demand for our specialized cloud infrastructure and services."

Financial services demand. CoreWeave said it inked big bank deals with banking giants Morgan Stanley and Goldman Sachs for proprietary trading.

Healthcare scaling AI. "We're also seeing significant growth from healthcare and life science verticals, and are proud of our partnership with customers like Hippocratic AI who built safe and secure AI agents to enable better healthcare outcomes," said Intrator.

Training and inference. Intrator said CoreWeave is working with customers to easily transition between training and inference. "We're helping these customers redefine how data is consumed and utilized globally as their critical innovation partner, and we are being rewarded for our efforts," said Intrator. "As they shift additional spend to our platform, we continue to execute and invest aggressively in our platform, up and down the stack to deliver the bleeding edge AI cloud services, performance and reliability that our customers require to power their AI innovations."

Intrator said:

"We really build our infrastructure to be fungible, to be able to be moved back and forth seamlessly between training and inference. Our intention is to build AI infrastructure, not training infrastructure, not inference infrastructure."

Storage workloads. CoreWeave said it is gaining storage share for AI-centric workloads. "Customers are shipping petabytes of their core storage to core week in the form of multiyear contracts. We are providing support for additional third party storage systems," said Intrator.

Bleeding edge expansion. Intrator said customers are focused on the latest hardware--Nvidia systems--to remain on the bleeding edge. "Clients are purchasing hardware that is appropriately state of the art for their use case. And as new hardware comes out, as new hardware architectures are released, they tend to come back in and purchase the same top tier infrastructure their next renewal," said Intrator.

Power and supply constraints. Intrator said the AI infrastructure market is "structurally constrained." "It is a market that is really working hard to try and balance and there are fundamental constraints at the power shell through the grid to the supply chains that exist within the GPUs to the mid voltage transformers," said Intrator. "There are a lot of different pieces that are constrained, but ultimately the most significant challenge right now is accessing power shells."