{"id":337226,"date":"2026-05-19T15:28:39","date_gmt":"2026-05-19T09:58:39","guid":{"rendered":"https:\/\/ebiztoday.news\/?p=337226"},"modified":"2026-05-19T15:28:40","modified_gmt":"2026-05-19T09:58:40","slug":"fully-autonomous-ai-demands-a-brand-new-enterprise-anchor","status":"publish","type":"post","link":"https:\/\/ebiztoday.news\/index.php\/2026\/05\/19\/fully-autonomous-ai-demands-a-brand-new-enterprise-anchor\/","title":{"rendered":"Fully autonomous AI demands a brand new enterprise anchor"},"content":{"rendered":"
\n

Fully autonomous AI has crossed the edge from concept to industrial reality, forcing enterprises to rethink the very foundation on which they run their businesses.<\/p>\n

The shift is arriving faster than most organizations anticipated, bringing with it an urgent latest set of economic and architectural demands. However the pressure is particularly acute for enterprises that also treat legacy infrastructure as the middle of their technology universe, in response to John Roese<\/a> (pictured), global chief technology officer and chief AI officer at Dell Technologies Inc.\u00a0Now, as agentic AI moves from buzzword to board-level priority, the conversation has gone from possibility to implementation \u2014 and price.<\/p>\n

\u201cI used to be introducing people to the word \u2018agentic\u2019 a few 12 months ago,\u201d Roese said. \u201cNow we\u2019ve realized that this concept of fully autonomous AI systems \u2014 of really shifting work into the machine layer \u2014 is now very real. At the identical time \u2026 the word \u2018tokennomics\u2019 is now in our vernacular because we\u2019ve realized that once you put this stuff into production at scale, the associated fee of various patterns of what you deploy are incredibly variable. Having a wise, intelligent method to use your hybrid infrastructure, to place the suitable workload in the suitable place to make use of the suitable model, is definitely not only a nice-to-have \u2014 it\u2019s required.\u201d<\/p>\n

Roese spoke with theCUBE\u2019s John Furrier<\/a> and Dave Vellante<\/a> at Dell Technologies World 2026<\/a>, during an exclusive broadcast on theCUBE, SiliconANGLE Media\u2019s livestreaming studio. They discussed fully autonomous AI infrastructure, the economics of hybrid deployment and the emergence of AI-centric engineering as a brand new organizational discipline. (* Disclosure below.)<\/em><\/p>\n

Fully autonomous AI demands a brand new enterprise anchor point<\/h3>\n

Roese has consistently suggested that success on this era won\u2019t come from chasing every AI breakthrough, but from constructing infrastructure able to keeping pace<\/a> \u2014 a view that puts the corporate squarely at odds with how most enterprises are currently approaching the issue. The legacy-versus-AI debate is one enterprises are increasingly getting unsuitable, with many attempting to bolt AI capabilities onto existing legacy, or \u201cbrownfield,\u201d environments, he added.<\/p>\n

\u201cThe most important impediment to navigating the legacy environment into the AI era is this error of considering you\u2019re doing it by having the legacy environment be the anchor,\u201d he said. \u201cThe one way it really works is [if] you truly have an AI environment. You may have to have a spot to construct agents and run them autonomously. You may have to have a security architecture that works with agentic. The brownfield becomes a feeder system to the brand new AI environment, but in the event you don\u2019t have the AI environment in any respect, you\u2019re just putting some AI features around a brownfield.\u201d<\/p>\n

That infrastructure philosophy extends directly into data. The\u00a0Dell AI Data Platform is an example of what it means to treat data not as a separate concern but as an integral component of the AI factory itself, in response to Roese. The toughest work in any advanced agent deployment has been data plumbing, specifically converting institutional knowledge into knowledge graphs and ontologies that agents can actually use.<\/p>\n

\u201cAn agent that just has access to a big language model will not be an enterprise agent,\u201d he said. \u201cAn enterprise agent also has access to a knowledge graph that organizes the proprietary enterprise knowledge. For those who give an agent each of those, and the knowledge graph is ground truth, it behaves higher, it’s more relevant, it’s more aligned to your small business.\u201d<\/p>\n

As enterprises mature those agentic deployments, the query shifts from whether agents can do the work as to whether they’ll do it reliably. Fairly than allowing agents to improvise, Dell has embedded digitized process knowledge directly into its agentic systems \u2014 giving them an outlined playbook to attract from, Roese explained. The excellence matters because unpredictable agents, nonetheless capable, can’t be trusted with the workflows that truly run businesses.<\/p>\n

\u201cWe’re big believers in digitized process,\u201d he said. \u201cWithin the old world, that digitized process led humans around. In the brand new world, once we construct agents, one in every of the tools they use to perform work is that they go to our digitized processes. The worth of giving agents more details about what the method is \u2014 how they need to work, somewhat than letting them invent \u2014 just makes them more predictable and more scalable.\u201d<\/p>\n

Stay tuned for the whole video interview, a part of SiliconANGLE\u2019s and theCUBE\u2019s coverage of Dell Technologies World 2026<\/a>.<\/p>\n

(* Disclosure: TheCUBE is a paid media partner for Dell Technologies World 2026. Sponsors of theCUBE\u2019s event coverage wouldn’t have editorial control over content on theCUBE or SiliconANGLE.)<\/em><\/p>\n

Photo: SiliconANGLE<\/h5>\n
\n
\n

Support our mission to maintain content open and free by engaging with theCUBE community. Join theCUBE\u2019s Alumni Trust Network<\/strong>, where technology leaders connect, share intelligence and create opportunities.<\/p>\n