Analyst(s): Brad Shimmin, Nick Patience
Publication Date: May 20, 2026
What is Covered in This Article:
- Dell introduced Deskside Agentic AI, enabling organizations to develop and run autonomous agents locally, reducing reliance on public cloud application programming interfaces.
- The company expanded its AI Data Platform with a new Data Orchestration Engine to index unstructured files and forge governed data pipelines.
- A strategic partnership with Starburst brings GPU-accelerated SQL analytics to enterprise environments, sharply improving query performance for data-heavy applications.
- Dell launched PowerRack, a fully integrated compute, networking, and storage system built to handle massive high-performance computing requirements.
- Advancing its hybrid deployment strategy, Dell prioritizes data gravity to give enterprises a predictable economic model for scaling agentic architectures.
- Dell announced a series of on-premises partnerships with frontier models, including Google, OpenAI, SpaceX/xAI, and Palantir.
The Event — Major Themes & Vendor Moves: Dell Technologies unveiled a major expansion of the Dell AI Factory with NVIDIA, aiming to push enterprise artificial intelligence out of the experimental phase and into full-scale production. The launch introduces the Dell Deskside Agentic AI solution, which pairs high-performance workstations with NVIDIA NemoClaw software to run autonomous agents locally. Further fortifying its infrastructure stack, Dell rolled out PowerRack, a turnkey rack-scale solution featuring integrated thermal and power management. Dell also announced a series of frontier model on-premises partnerships, bringing Google Gemini 3 (on PowerEdge XE9780 in a confidential computing environment), OpenAI GPT-5.5 and Codex, SpaceX/xAI’s multimodal capabilities, and Palantir Foundry’s ontology layer – deployed on ObjectScale and PowerFlex – to the AI Factory stack. Additional open-source models, including DeepSeek variants, Minimax, GLM, and Kimi, are available through the Dell Enterprise Hub on Hugging Face. The company also upgraded its AI Data Platform with a new Data Orchestration Engine and highlighted a strategic partnership with Starburst. Ultimately, Dell’s strategy champions hybrid deployments and an open ecosystem to tame rising inference costs and drive concrete business outcomes.
Is the Cloud Too Expensive for Agentic AI? Dell Bets on Localized Tokens
Analyst Take: Enterprise artificial intelligence deployments face mounting economic and architectural constraints. Many companies assume public cloud application programming interfaces will seamlessly scale to handle the complex, multi-step workflows of autonomous agents, overlooking harsh operational realities. Powering continuous agentic reasoning demands massive token consumption, which triggers severe cost volatility and exposes proprietary corporate data to unnecessary network friction.
Enterprises are rapidly abandoning isolated conversational interfaces in favor of physical AI and industrial autonomy. This transition demands robust infrastructure capable of sustaining autonomous digital workers. According to the Futurum Research 2026 Key Issues & Predictions report, 71% of CIOs are currently reevaluating cloud workload placement, driven by pressing concerns over AI cost structures and data gravity. Dell is counting on this reevaluation. By bringing advanced models directly into the corporate environment, Dell is allowing organizations to bypass the unpredictable economics of metered intelligence altogether.
The Agentic Deskside and Localized Execution
Our State of the Market Report: Data Intelligence, Analytics, and Infrastructure, Q2 2026 indicates that major vendors are chasing data gravity, enabling AI agents to operate directly where data resides to minimize latency and tighten security. Dell hopes to tackle this directly with its Deskside Agentic AI solution. Workgroups can now deploy dedicated systems powered by the NVIDIA NemoClaw reference stack to handle vital models ranging from 30 billion to 1 trillion parameters.
Building and deploying agents right next to the data source enforces strict governance over intellectual property and neutralizes the bandwidth expenses of transmitting massive datasets across wide-area networks. Organizations leveraging this localized infrastructure can achieve a break-even point against public cloud operational costs in the short term and potentially slash overall spend compared to public cloud APIs over the long term. This secure sandbox frees software engineering teams and researchers to iterate endlessly without racking up exorbitant utility bills.
Redesigning the Data Platform Through Strategic Partnerships
Data orchestration remains the fundamental bottleneck for autonomous agents. Fragmented silos effectively blind AI systems, blunting the impact of even the most advanced frontier models. As infrastructure evolves from passive repositories to active storage architectures that embed real-time data discovery and vector acceleration, vendors must adapt. Dell intends to help here through several updates to its AI Data Platform, which confronts this limitation head-on with a new Data Orchestration Engine that delivers up to 12x faster vector indexing to reliably prepare and route critical information, according to Dell.
Monolithic data platforms are steadily fracturing into what Futurum would classify as a composable data intelligence stack governed by open standards such as Apache Iceberg. Dell navigates this fragmentation through highly strategic ecosystem integrations that preserve client flexibility. The company’s collaboration with Starburst exemplifies this open framework perfectly. Instead of forcing customers into a proprietary analytical repository, Dell uses the Starburst engine to deliver GPU-accelerated SQL analytics across both structured and unstructured data. According to Dell, this integration boosts query performance by up to 6x on NVIDIA Blackwell graphics processing units, rapidly accelerating the insights required to run data-intensive agentic applications.
Furthermore, Dell and NVIDIA have built a deeply integrated environment that transcends basic hardware aggregation and software integration. Embedding NVIDIA OpenShell, for example, creates a secure runtime spanning the entire hardware portfolio, ranging from individual workstation towers up to massive PowerEdge XE server racks. With this foundation, organizations can develop agents within a strict policy-enforcement layer. This robust architecture prevents unauthorized access while allowing teams to build and test complex multi-agent workflows using deployable blueprints such as AI-Q 2.0.
The Future of Abundant Intelligence
During his keynote address, Michael Dell offered a compelling vision regarding the normalization of computing power, observing that abundant intelligence has arrived and is swiftly becoming foundational infrastructure. He drew a striking parallel: electricity truly transformed the world when it left the power plant, and AI will reshape our reality as it moves beyond the screen.
Pushing raw processing power into physical spaces—from ambulances to bustling factory floors—demands localized, highly secure, and low-latency hardware. The barrier separating imagination from execution is rapidly collapsing. Organizations will generate vastly more software as they codify daily workflows into autonomous systems. To support this massive operational expansion without buckling under financial strain, enterprise leaders must rethink their heavy reliance on public cloud architectures. Adopting a robust hybrid deployment model allows businesses to treat intelligence as a predictable core utility rather than a volatile operational expense.
Frontier Model Partnerships: Dell’s Challenge to Hyperscaler Lock-In
Some of the most strategically significant announcements at DTW 2026 were not hardware; they were on-premises partnerships with the frontier model. Dell now offers enterprises the ability to run Google Gemini 3, OpenAI Codex, SpaceX/xAI, and Palantir Foundry within their own infrastructure, with data remaining on-premises and model access decoupled from any single cloud provider’s commercial relationship.
This directly challenges the hyperscaler bundling model, in which access to frontier models is effectively tied to cloud infrastructure consumption. Dell’s argument – that the differentiator in enterprise AI is not which model you run but where your data lives and how you govern it – has been a consistent positioning theme, but it now has substantive ecosystem support behind it. The confidential computing environment for Google Gemini 3 is particularly relevant: running frontier models in a hardware-enforced confidential computing environment, where even the infrastructure operator cannot access model weights or inference data, is a prerequisite for regulated-industry production deployments in financial services, healthcare, and government.
The 2026 Futurum AI Decision Makers Survey shows that data control and cost predictability are now the primary concerns for enterprises moving AI workloads from pilot to production – ahead of model performance. Dell’s hybrid AI positioning maps directly onto those priorities.
What to Watch:
- The stakes are high for Dell, as the company is endeavoring to solve a very difficult problem. For example, the reliability of unsupervised agents remains a significant hurdle. Organizations will require robust auditing tools to verify the exact reasoning paths of digital workers before handing over transaction authority.
- Watch how the broader cyber industry responds to Dell’s partner-led approach. Established markets like identity and access management warrant particular attention. Coordinating multiple autonomous agents demands comprehensive identity and access governance protocols to prevent unauthorized data exposure across disparate corporate systems.
- Integrating advanced rack-scale hardware into existing network topologies poses a significant structural challenge. How will Dell help its enterprise customers modernize core switching and routing architectures to support massive east-west traffic flows?
- The Google Gemini 3, OpenAI Codex, and xAI on-premises partnerships were announced as available or near-available. Watch how quickly these translate from announcement to documented enterprise production deployments, specifically in regulated industries where the confidential computing environment is a prerequisite rather than an option.
See the complete press release regarding the AI Factory expansion on the Dell website.
Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.
Other Insights From Futurum:
Precision Over Prose: Why SAP Knowledge Graph is the Secret to Production-Ready AI
Microsoft Agent 365 Turns Shadow AI Into a Governed Asset Class
Can ServiceNow’s Autonomous AI Foundation Finally End the Enterprise ETL Tax?
Author Information
Brad Shimmin is Vice President and Practice Lead, Data Intelligence, Analytics, & Infrastructure at Futurum. He provides strategic direction and market analysis to help organizations maximize their investments in data and analytics. Currently, Brad is focused on helping companies establish an AI-first data strategy.
With over 30 years of experience in enterprise IT and emerging technologies, Brad is a distinguished thought leader specializing in data, analytics, artificial intelligence, and enterprise software development. Consulting with Fortune 100 vendors, Brad specializes in industry thought leadership, worldwide market analysis, client development, and strategic advisory services.
Brad earned his Bachelor of Arts from Utah State University, where he graduated Magna Cum Laude. Brad lives in Longmeadow, MA, with his beautiful wife and far too many LEGO sets.
Nick Patience is VP and Practice Lead for AI Platforms at The Futurum Group. Nick is a thought leader on AI development, deployment, and adoption - an area he has researched for 25 years. Before Futurum, Nick was a Managing Analyst with S&P Global Market Intelligence, responsible for 451 Research’s coverage of Data, AI, Analytics, Information Security, and Risk. Nick became part of S&P Global through its 2019 acquisition of 451 Research, a pioneering analyst firm that Nick co-founded in 1999. He is a sought-after speaker and advisor, known for his expertise in the drivers of AI adoption, industry use cases, and the infrastructure behind its development and deployment. Nick also spent three years as a product marketing lead at Recommind (now part of OpenText), a machine learning-driven eDiscovery software company. Nick is based in London.
