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

When a technology is evolving as fast as artificial intelligence, CFOs and the finance department struggle to crowbar AI infrastructure investments into traditional depreciation models.

Typically, enterprises depreciate capital spending on IT infrastructure over multiple years. For instance, servers have a useful life of 7 years, according to the IRS. Hyperscale cloud companies tend to tweak the useful life of a server. For instance, Meta in its annual report said certain servers and network assets had a useful life of 5.5 years for fiscal 2025. Amazon now reckons the useful life of some of its servers is now 5 years down from 6 years a year ago. Alphabet depreciates over 6 years.

Here's the catch: 5 years in AI infrastructure is a lifetime. Nvidia now has an annual cadence for its GPUs and provides visibility into the roadmap through 2027. CEO Jensen Huang's bet is that Nvidia customers will continue to invest in the latest and greatest accelerated computing infrastructure because there's value and competitive advantage.

• Nvidia launches Blackwell Ultra, Dynamo. outlines roadmap through 2027
• Nvidia GTC 2025: Six lingering questions

Do companies go with shorter depreciation cycles, lease gear or just provision from the cloud even though operating expenses are already stretched? As Nvidia extends more into the enterprise and industry applications, this financial planning conundrum is going to go mainstream. It's no wonder that Nvidia has teams focused on financial solutions.

With that in mind, Nvidia held a virtual panel at GTC 2025 focused on these accounting issues. What does traditional planning look like when AI is advancing too quickly for typical technology refresh cycles?

Bill Mayo, SVP for research IT at Bristol Myers Squibb, has been investing in AI and machine learning for more than a decade, but AI advances today are moving faster than ever. "The challenge is we've had this probably first and second derivative improvement in in the pace of change that has completely broken financial models up and down the stack," said Mayo.

Richard Zoontjens, lead of the supercomputing center at Eindhoven University of Technology, said "we really need compute to compete in this era and that means financial systems have to support this fast moving world."

Zoontjens noted that the current AI cycle doesn't fit in with typical depreciation schedules and financial models. "If you buy new tools every five years, well you're not competing anymore. After two years, you lose talent and you lose innovation," he said.

The solution is that financial modeling will have to move faster. Mayo noted that Bristol Myers Squibb (BMS) has seen rapid improvement in compute, but there's still not enough to reach its biology vision. "At its core, biology is computation," said Mayo.

Mayo said that if you're using today's tech stack five years from now you're behind. Mayo said BMS hasn't figured out the financial model behind AI investments yet, but did say that its first swing was to move to a four-year depreciation cycle.

Zoontjens said his group opted for a flexible two-year renewal cycle. Flexibility is key. Zoontjens said sometimes his supercomputing center can stretch the system, but sometimes has to upgrade faster.

"The system and the lifecycle management contract that we have now provides that flexibility," said Zoontjens. "It gives us control and better agility to move and stay state of the art."

Mayo said AI investment and modeling has to revolve around the patient population and have the best insights to improve lives. That alignment helps with the costs, but BMS AI infrastructure is an operating expense due to cloud delivery. The problem is that cloud provider demand for AI infrastructure is high. Mayo said:

"We can't afford to buy a new (Nvidia) Super Pod every year and just use it for a year. I can't afford it at an OPEX rate, and frankly, neither can hyperscalers afford to buy enough to make it available fast enough that we can all consume that way."

The current situation may indicate that the havoc hitting financial models may be transitory, said Mayo, who said on-prem, co-located infrastructure as well as cloud AI services are all in the mix. "The fact of matter is, I'm buying through a time window that maybe three years from now, the financing problem might have solved itself, but TBD on that," said Mayo.

Here are a few themes on financing AI infrastructure from the IT buyers and Nvidia's Timonthy Shockley, global sales at Nvidia Financial Solutions, and Ingemar Lanevi, head of Nvidia Capital:

• Plan for data center investments that incorporate power and space savings. Nvidia systems have needed less space with each new system.
• Plan for more agile upgrade cycles to maintain capacity to compete in industries.
• There's no right answer that covers all the financial bases so there will be a mix of cloud and on-prem decisions to be made.
• Long-term depreciation will be an issue for the foreseeable future.
• Long-term cloud contracts and leases can be a challenge.
• Cross-functional teams will have to make financing decisions based on what needs to be achieved now and then where things will move later.
• Leasing models may make sense for AI infrastructure at the moment for cash flow purposes and building in upgrades.
• It's possible that a secondary market for AI infrastructure emerges for what Mayo called "gently used Super Pods." The accelerated computing market is young relative to CPUs so a secondary market may take time.
• Enterprises may look to monetize remaining residual value of AI infrastructure when it's not helpful to the buyer anymore.
• Segment investments for what needs to be cutting edge and adjust the financial model accordingly. Non-cutting edge tech investments can be depreciated over a longer period.
• Today's AI infrastructure spend is governed by financial systems, but may have to flip in the future to account for product cycles.

Mayo added the disconnect between financial planning and the AI opportunity is just a place in time.

"It's going to get solved. There's a right answer for the use case or the situation you're grappling with right now. Maybe it's a funding model, maybe it's a cash constraint. As long as we're open to try whatever, we're going to solve this problem, and then we're going to use this solution to solve all the other problems."

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

Agentic AI is likely to have an unwelcome side effect: Data wars and lawsuits between platforms.

Process mining company Celonis filed a US antitrust complaint against SAP in the US District Court of California, San Francisco Division. The complaint alleges that SAP is restricting Celonis' access to data on its ERP platform in favor of SAP-owned rival Signavio.

Celonis and SAP have had a long relationship and the former was in the SAP Startup Focus program in 2012. Celonis alleged in its complaint that "SAP is leveraging its control over its ERP ecosystem and the impending forced migration of customers to SAP's S/4HANA cloud-based ERP solution to prevent SAP customers from sharing their own data with third-party providers, including Celonis, without paying prohibitively expensive fees.

The complaint further argues that SAP is bundling Signavio and preventing competitors like Celonis from extracting data from a customer's ERP system. Celonis alleges that SAP is making it impossible to use non-SAP process mining. SAP hasn’t responded to the lawsuit yet.

While Celonis is connecting to SAP for process intelligence, you can easily imagine that these data access skirmishes will proliferate as millions of AI agents start trying to complete tasks autonomously. Data ecosystems work--until they don't. Is orchestrating processes as an independent third party really that different from the emerging AI agent orchestration platforms?

There will be no shortage of neutral-ish agentic AI platforms. ServiceNow's latest release of its Now Platform has a bevy of tools to connect agents and orchestrate them. Boomi launched AI Studio. Kore.ai launched its AI agent platform, and eyes orchestration. Zoom evolved AI Companion with agentic AI features and plans to connect to other agents. Without data connections there won’t be cross platform orchestration.

Meanwhile, enterprise technology giants are all playing the platform game. When your data is on a platform you're locked in to some degree. For agentic AI to meet its promise, enterprises will need to connect agents on multiple platforms. It's not difficult to envision a world where SAP agents connect with Salesforce Agentforce and perhaps, ServiceNow, AWS and Google Cloud to autonomously execute a task.

Beyond the standards of AI agent communications there will be data handoffs. Some platforms, say Salesforce and Google Cloud or Databricks and SAP, will have zero copy data sharing arrangements. Other vendors may collect tolls from potential rivals not in an alliance. These tolls are going to the basis of future data wars.

• Enterprise AI: Here are the trends to know right now
• AI agents bring consumption models to SaaS: Goldilocks or headache?
• Agentic AI: Three themes to watch for 2025
• Agentic AI without process optimization, orchestration will flop

None of these data skirmishes are surprising. Enterprise vendors have typically tried to keep your data and prevent rivals from benefiting. CIOs will tell you tales about full use vs. restricted use when it comes to extracting data from one platform and sharing it with third party applications. Procurement departments can rant about expensive API usage fees and rate limits. Tight integration is also designed to keep you on one platform and make it difficult to connect third party vendors. Hyperscale cloud vendors have played the data egress fee game.

Sure, you can connect your data to third parties and even transfer data to new platforms. Just expect some technical and financial pain.

Where the enterprise technology data playbook becomes worrisome is agentic AI. The customer should have control of the enterprise data. Enterprises should be able to connect platforms and their agents to complete processes and autonomously make decisions. Vendors today like to pretend that agentic AI will only happen on their platforms.

It's worth noting that agentic AI is going to tax transactional systems with API calls. There is SaaS scale in terms of API calls and then there's agentic AI calls, which will be a whole new ballgame. There may be a need for some pricing model to address the strain.

Either way, agentic AI is going to scale the business-as-usual data charges and the big platforms are naturally inclined to put up a few barriers to smaller vendors. Simply put, Celonis' lawsuit vs. SAP isn't likely to be a process mining one off. Agentic AI will require the same type of data access at scale. The data wars between vendors is just starting and customers are going to be caught in the middle. Just remember your data belongs to you and not your vendor.

Relevant research:

• AI Trends for 2025 and Beyond
• Behind the Scenes: The Force Behind Agentforce
• There Is Only ‘One’ (1) IT Strategy in 2025 and Beyond
• Exponential Efficiency in the Age of AI

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

Heather Wilson, Senior Vice President and Chief Communications and Marketing Officer at AccentCare, sits at the intersection of customer experience, employee experience, marketing and talent recruitment.

Why the multiple roles? For Wilson and AccentCare they all blend together. AccentCare is one of the largest post-acute healthcare providers in the US covering the range of personal care to clinical services like nursing and physical therapy as well as hospice care.

• Will CX suffer from an overreliance on AI?
• Constellation Research Ambient Experience Summit 2025: 10 CX takeaways on people, data, AI
• AX100 Spotlight: RentSpree's Michael Storey on the intersection of CX, AI and success metrics

The company's ability to tell a good human story and drive experiences also attracts talent amid a healthcare worker shortage. I caught up with Wilson at Constellation Research's Ambient Experience Summit. Here are some of the takeaways from our chat and the questions that followed.

Why customer experience (CX) and employee experience (EX) is critical to healthcare. "My role has both the CX and EX experience because, while our end client is physicians and healthcare systems, we also have to ensure a good experience for patients and families—they are a secondary referral source," she said.

The need for narratives. Wilson was brought on at AccentCare to tell the stories of clinicians and patients. She continually emphasizes storytelling in internal meetings, marketing and executive communications. "Storytelling is in my DNA. We have amazing stories in the everyday work that our clinicians do, but they weren’t being told," said Wilson. "Everything we do starts with a patient story. When executives speak, when they visit branches—we make sure they have real patient stories because it reminds us why we do this."

Transformation. Wilson said transformation in healthcare is often slow due to technology, legacy systems and older processes. Wilson modernized AccentCare's digital marketing strategy, but noted that "we live in an industry where paper brochures are a thing."

AI and analytics. Wilson said AI and analytics is being used at AccentCare for sales enablement, marketing targeting, and talent acquisition and recruitment. "We have to blend the traditional with the digital," said Wilson.

Digital transformation challenges. Wilson said barriers to healthcare digital transformation is hampered by the lack of interoperability between Electronic health records, regulations on patient data and aging patient demographics that keep non-digital engagement methods going. Full digital transformation is likely 10 years away at least, said Wilson. "Healthcare is still running on 1990s systems, and they’ve gotten away with it for a long time. But with AI, the pressure is on to finally modernize," she said.

The intersection of CX and EX. Wilson said word of mouth is critical to recruiting and patient care quality. "If you’re a clinician, you want to work at a place known for top-quality care. If you’re a doctor, you want to refer patients to a company you trust," said Wilson. "If we don’t have enough nurses, we can’t grow. It’s that simple."

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

Accenture said it may start seeing the effects of a US government funding pause and enterprises growing cautious amid geopolitical and economic volatility.

Speaking on Accenture's second quarter earnings conference call, CEO Julie Sweet laid out the current conditions for the consulting giant. Accenture Federal Services accounted for 8% of the company's global revenue and 16% of Americas revenue in fiscal 2024.

She said:

"As you know, the new administration has a clear goal to run the Federal government more efficiently. During this process, many new procurement actions have slowed, which is negatively impacting our sales and revenue. In addition, recently, the General Service Administration has instructed all federal agencies to review their contracts with the top 10 highest paid consulting firms contracting with the U.S. government, which includes Accenture Federal Services.

The GSA's guidance would determinate contracts that are not deemed mission critical by the federal -- by the relevant federal agencies. While we continue to believe our work for federal clients is mission critical, we anticipate ongoing uncertainty as the government's priorities evolve and these assessments unfold."

That caution has spilled over to enterprises, which have gone from bullish at the end of 2024 to uncertain. Other vendors have also noted that enterprises have turned cautious in recent weeks.

Sweet said enterprises have only recently turned cautious and haven't paused spending just yet. See: AI? Whatever. It's all about the first party data

"We are seeing an elevated level of what was already significant uncertainty in the global economic and geopolitical environment, marking a shift from our first quarter FY ‘25 earnings report in December," said Sweet. "Businesses are trying to process what this (volatility) might mean."

The forecast from Accenture lands as its results for the second quarter were solid excluding new bookings growth.

For the second quarter, Accenture reported earnings of $2.82 a share on revenue of $16.7 billion, up 5%. Both top and bottom lines beat estimates. Accenture saw $1.4 billion in generative AI bookings in the quarter.

The issue for Accenture is that new bookings for the second quarter were $20.91 billion, down 3% from a year ago. New bookings growth in the first quarter was only 1%. That slowdown in new bookings may indicate a great pause amid uncertain economic conditions.

As for the outlook, Accenture said its third quarter revenue will be between $16.9 billion and $17.5 billion compared to estimates of $17.22 billion. For fiscal 2025, Accenture projected revenue growth of 55 to 7% compared to 4% to 7% last quarter.

AI projects to save the day?

Sweet said Accenture is still seeing strong demand for AI services. Enterprises are focused on their "digital core with more AI being built-in."

"For our clients, the twin themes of achieving both cost efficiency and growth continue. The number of clients embracing Gen AI is increasing significantly and we are starting to see some tangible examples of scale in data and AI," said Sweet.

Accenture is also scaling its AI Refinery platform that focuses on business processes and agentic AI. Sweet cited customer wins in manufacturing, automotive, food and telecom via a partnership with Telstra, a leading telecom in Australia.

• Enterprise AI: Here are the trends to know right now

Sweet added that Accenture is using its AI Refinery platform and automation to cut costs with AI.

In the end, Accenture may be seeing a tale of two businesses. AI and data projects will get funding and everything else is downgraded. Sweet said the US geopolitical and economic picture is uncertain and it's unclear how customers will react, but spending in Europe could pick up. Sweet said:

"Everyone is well aware that in the last few weeks, there's been an elevated level of what was already significant uncertainty and there's a couple of big themes around that, obviously tariffs, and that's a global discussion that is not just an Americas discussion.

And there’s consumer sentiment, which is a little bit more of an Americas discussion. We're already in the heart of the discussions of clients globally who are talking about it."

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 launched Oracle AI Agent Studio for Fusion Applications in a move designed to enable enterprises to customize AI agents across its platform.

The move, announced at Oracle Cloudworld in London, is the latest in a series of vendor announcements aimed at creating, deploying and orchestrating AI agents.

There is not shortage of AI agent platforms. ServiceNow's latest release of its Now Platform has a bevy of tools to connect agents and orchestrate them. Boomi launched AI Studio. Kore.ai launched its AI agent platform, and eyes orchestration. Zoom evolved AI Companion with agentic AI features and plans to connect to other agents. Salesforce obviously has Agentforce.

According to Oracle, AI Agent Studio will enable customers to create and manage their own AI agents. Oracle already has more than 50 AI agents embedded into its Fusion applications.

Oracle AI Agent Studio is available at no additional cost and includes testing, validation and security to create agents within Oracle Fusion Applications.

Holger Mueller, analyst at Constellation Research said:

"The new Oracle AI Agent Studio is an impressive next step for Oracle's AI strategy. To truly optimize the impact of AI agents, organizations need to be able to customize the way they work to fit their unique business needs. The evolution of AI across the enterprise is moving at a rapid pace and by enabling agents to be created, extended, deployed, and managed across the entire enterprise, Oracle will help its customers accelerate adoption and automation.

It's good to see Oracle delivering on a consistent architecture strategy for its AI platform and Fusion Application portfolio as AI can't be built overnight. Oracle's vertical stack depth – from the cloud infrastructure over the database makes this a compelling offering for its customers."

Oracle AI Agent Studio includes:

• Agent pre-built templates.
• Orchestration tools and agent extensibility with documents, tools, prompts and APIs.
• LLM choice via models optimized for Fusion via Llama and Cohere or third-party options.
• Native integration with Fusion tools.
• Connections to third party agents.

Oracle Database@Azure expands

Separately, Oracle said Oracle Exadata Database Service on Exascale Infrastructure on Oracle Database@Azure is generally available.

Oracle also said Oracle Base Database Service on Oracle Database@Azure will be available soon.

The company also said Oracle Database@Azure is available in the Microsoft Azure Ease US 2 region and now available in 14 regions. Oracle plans to roll out 18 regions on Azure in the next 12 months.

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

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

Softbank said it will acquire chipmaker Ampere Computing in a $6.5 billion all-cash deal. Oracle and Carlyle, Ampere's primary investors, have agreed to sell their stakes.

Carlyle owned 59.65% of Ampere, Oracle owned 32.27% and Arm owned 8.08%. Arm is a subsidiary of Softbank Group and Ampere is a licensee of Arm.

According to Softbank, Ampere, founded in 2018, will operate as a wholly owned subsidiary of the company. Softbank has been doubling down on AI infrastructure and has invested in the Stargate project and Cristal Intelligence in partnership with OpenAI and has venture funding spread across multiple startups.

Ampere designs high-performance, energy efficient processors for cloud and AI workloads. Softbank said: "Ampere's expertise in developing and taping out ARM-based chips can be integrated, complementing design strengths of Arm Holdings."

Softbank CEO Masayoshi Son said in a statement that Ampere accelerates its vision for "Artificial Super Intelligence" and "deepens our commitment to AI innovation in the United States."

Ampere's processors are Altra, Altra Max and Ampere One family. The company has focused on cloud-native energy efficient processing, but has pivoted to AI workloads too.

• Softbank Corp. first to land Nvidia Blackwell systems in bid to be genAI provider in Japan
• OpenAI's Stargate Project diversifies its cloud infrastructure
• GenAI's 2025 disconnect: The buildout, business value, user adoption and CxOs
• Qualcomm, Arm cheer cheaper models, AI inference at the edge

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

Nvidia's GTC conference kicked off with a long keynote from CEO Jensen Huang, a roadmap extending in 2028 and an integrated AI stack that's hard for rivals to match.

Here's a look at the questions that are lingering after GTC kicked off. 

Can Nvidia's cadence keep demand going?

Nvidia's ability to cannibalize itself with an annual cadence and dangle enough value and performance to convince customers to upgrade has been impressive.

What's unclear is whether this roadmap can keep being Nvidia's greatest trick. Nvidia CEO Jensen Huang joked during his keynote that his salespeople aren't going to be happy that he keeps dissing Hopper, the GPU that arguably started an AI boom. But Blackwell is way better. Blackwell Ultra will be better than Blackwell. Vera Rubin, Rubin Ultra and Feynman will all be better than what was there a year before.

Huang's bet is that AI will lead to a continuing scale up and scale out cycle for AI factories. Once you scale up, scaling out will lead to better cost of ownership. "Rubin will bring costs down dramatically," said Huang.

GTC 2025:

• Nvidia launches Blackwell Ultra, Dynamo. outlines roadmap through 2027
• Oracle Cloud adds Nvidia AI Enterprise, Nvidia Blackwell GB200 NVL72
• Nvidia launches DGX Spark, DGX Station personal AI supercomputers
• Nvidia's model parade: Llama Nemotron, Cosmos additions, Isaac GROOT N1

Here's the catch: Agentic AI will lead to more AI infrastructure. Cheaper models will bring more consumption as will enterprise use cases. The wrinkle is that Nvidia's big customers--AWS, Microsoft Azure, Google Cloud and Meta--all are building custom silicon to lessen their dependence on Nvidia. Can hyperscalers catch up and do they even half to if good enough AI infrastructure becomes the norm? Nvidia's answer is that it can deliver performance and value faster.

Can DeepSeek and cheaper models add to Nvidia's moat?

Nvidia's GTC opener revolved around reasoning models and ways to scale. There's a good reason for that--Wall Street is worried that reasoning models will lessen the need to spend on Nvidia gear.

The jury is way out on the DeepSeek impact, but I'd call the impact on Nvidia mostly a coin flip. Cheaper models may speed up enterprise usage and benefit Nvidia. Or cheaper models may mean good enough AI infrastructure means the latest GPU can wait.

• DeepSeek: What CxOs and enterprises need to know
• DeepSeek's real legacy: Shifting the AI conversation to returns, value, edge
• Nvidia strong Q4, outlook eases AI infrastructure spending fears for now

Will the AI factory vision become reality?

The short answer is that Nvidia's AI factory vision is going to be reality. The debate is over timing and whether there will be hiccups or overcapacity at some point.

Nvidia's roadmap is public and on a one-year rhythm because you need time to plan AI factories. You need energy, which is the gating factor for AI, as well as land and all of this stuff that goes beyond infrastructure.

Nvidia has a roadmap to Gigawatt AI factories. How fast that road gets paved remains to be seen.

Is Nvidia now that de facto enterprise infrastructure provider?

If you believe that AI will be at the center of every workload, it's a no-brainer to think that Nvidia will power most data centers. There's a reason that Nvidia has expanded so heavily into networking and even desktops. It wants to offer you the full stack.

It remains to be seen whether enterprises build out on-prem AI operations, but that's why Nvidia is also focused on software and open-sourcing models. Its Llama-based models tailored for industry use cases will be used by SAP, ServiceNow and others.

Whether the Nvidia stack becomes the enterprise stack remains to be seen, but I wouldn't rule it out. GM is betting on Nvidia for its AI factory and the industry references cited by Huang are impressive.

All you have to do is look at Nvidia's networking and storage plans to realize the company has more on its mind than GPUs. The key vendors in compute, storage and networking are all following Nvidia's lead.

How long until Nvidia's robotics vision becomes reality?

Huang spent a lot of time talking about models for robotics and the future.

Nvidia's bet is that there will be billions of digital workers to collaborate with humans, there will be a shortage of employees and robots will fill that gap. Robots are likely to be less expensive, but don't bet against a $50,000 annual cost.

There was some evidence that Nvidia's autonomous vehicle business was gaining traction in its most recent quarter. Robots--even the humanoid variety--may be here closer than you'd think due to models that can do a lot more than language. Watch Nvidia's physical AI push closely since it's the enabler for robotics going forward.

How underappreciated is Nvidia's software stack?

Yes, Nvidia pays the bills with accelerating computing systems, but its software stack is what maintains the company's dominance.

Aside from the bevy of models to advance various enterprise use cases, Nvidia Dynamo is a sleeper hit in GTC. Huang said Dynamo is the "operating system of the AI factory."

Dynamo separates the processing and generation phases of large language models on different GPUs. Nvidia said Dynamo optimizes each phase to be independent and maximize resources.

By breaking various AI workloads up and optimizing compute, Dynamo may become the enabler for Nvidia's entire stack. When running the DeepSeek-R1 model on a large cluster of GB200 NVL72 racks, Dynamo boosts the number of tokens by 30x per GPU.

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 and Nvidia expanded their partnership in a move that will bring Nvidia AI Enterprise, Nvidia Blackwell GB200 NVL72 and agentic AI blueprints to Oracle Cloud Infrastructure.

Constellation Research analyst Holger Mueller said the expanded partnership between the two companies makes sense.

"Oracle is working hard to become the premier place for enterprises to tap into Nvidia resources. The strategy is key further build out Oracle’s lead in OCI and transactional enterprises. And as Oracle has not announced any plans to build custom AI chips – they are a naturally preferred partner for Nvidia. Joint customers will welcome this announcement."

Indeed, Oracle's cloud infrastructure (OCI) is being used by enterprise and hyperscale customers for training. Details of the expanded partnership include the following:

• Oracle said that Nvidia AI Enterprise will be available natively in OCI Console. The move will reduce the time to deploy the service and provide direct billing and support. The OCI Console with Nvidia AI Enterprise will be available in Oracle's distributed cloud.
• OCI customers will have access to more than 160 tools for training and inference as well as Nvidia NIM microservices.
• OCI will be among the first cloud providers to offer customers the next-gen Nvidia Blackwell chips. Specifically, OCI is now offering Nvidia Blackwell GB200 NVL72 on OCI Supercluster with up to 131,072 Nvidia GPUs.

Larry Ellison, CTO of Oracle, teased the Nvidia-based supercluster when the company reported third quarter earnings. Ellison was touting Oracle's strong infrastructure as a service growth.

Ellison said:

"AI training and multi-cloud database are experiencing hyper growth. We are in the process of building a gigantic 64,000 GPU, liquid-cooled NVIDIA GB 200 cluster for AI training. Our multi-cloud business at Amazon, Google and Microsoft grew 200% in the last three months alone. But in addition to these rapidly growing existing businesses, new customers and new businesses are migrating to the Oracle Cloud at an unprecedented rate.

The capability we have is to build these huge AI clusters with technology that actually runs faster and more economically than our competitors."

Oracle said it is taking orders for its AI supercomputer with Nvidia Blackwell Ultra GB300 GPUs.

The two companies said they will enable vector embeddings and vector indexes in AI Vector Search workloads in Oracle Database 23ai using Nvidia GPUs.

OCI AI Blueprints will provide no-code deployment recipes without manually provisioning infrastructure. OCI AI Blueprints will reduce GPU onboarding time with hardware recommendations, Nvidia NIM and observability tools.

Nvidia NIM now in OCI Data Science. Nvidia NIM will be available directly in OCI Data Science for real-time AI inference use cases.

• Oracle Q3 a miss, but backlog portends big revenue growth ahead
• Oracle's cloud, AI plans are a master class in co-opetition
• OpenAI's Stargate Project diversifies its cloud infrastructure

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

Nvidia launched DGX Spark, formerly Project Digits, and DGX Stations as it aims to bring AI supercomputers to students, developers, researchers and data scientists.

Project Digits made a splash at CES 2025 and now Nvidia is making good on its expansion plans. Nvidia is hoping to enable users to develop and run models locally before uploading them to the cloud for production.

DGX Spark and DGX Station will run on Nvidia's Grace Blackwell architecture that powers data centers. Asus, Dell, HP and Lenovo will build DGX Spark and DGX Station devices.

Nvidia CEO Jensen Huang said DGX Spark and Station are a "new class of computers." "With these new DGX personal AI computers, AI can span from cloud services to desktop and edge applications," said Huang.

Here's a look at the details of the DGX systems.

DGX Spark

• DGX Spark runs on the Nvidia GB10 Grace Blackwell Superchip that features a Blackwell GPU, fifth-gen Tensor Cores and FP4 support.
• DGX Spark will deliver up to 1,000 trillion operations per second of AI
• compute for fine-tuning and inference with the latest AI reasoning models.
• The device will include models such as Nvidia Cosmos Reason and Nvidia GR00T N1.
• The GB10 Superchip uses NVIDIA NVLink-C2C interconnect technology to deliver a CPU+GPU-coherent memory model with 5x the bandwidth of fifth-generation PCIe.
• Reservations for DGX Spark are open at Nvidia's site.

DGX Station

• Nvidia's DGX Station is the first desktop built with the Nvidia GB300 Grace Blackwell Ultra Desktop Superchip.
• DGX Station has 784GB of coherent memory space.
• The device has Nvidia's ConnectX-8 SuperNIC with support for networking at up to 800Gb/s.
• DGX Station has Nvidia's CUDA-X AI platform, access to NIM microservices and AI Enterprise.
• DGX Station will be available from Asus, BOXX, Dell, HP, Lambda and Supermicro later this year.