Next in Tech

Just in time for Halloween, we're diving into headless AI. It's not a phantom agent roaming an old data center, but rather the next stage of application structures that research director Sheryl Kingstone discusses with host Eric Hanselman. A certain CEO predicted in 2024 that agentic AI would be the death of SaaS applications. The expectation was driven by the thought that agents would gather all of the needed information to run businesses and take the needed actions. It's an idea that missed the importance of not only the volumes of data that are housed in SaaS, but, more importantly, the context, in the form of business logic, controls and linkages, that is also embedded in these systems. Headless application architectures look to enable the promise of agentic AI.
Headless is not a new concept. We've been trying to link applications together for at least a decade, but there have been limitations to the various approaches, SOAP, XML data exchange, to name but a few. One of the major constraints has been the ability to translate context between systems. Moving data was reasonably simple. Expressing the context that surrounds that data is much more complex. With the Model Context Protocol (MCP), we've got an option to address that gap. It requires that IT teams step up a level of abstraction to manage data, becoming the custodians of this critical raw material.
 
More S&P Global Content:
Next in Tech | Ep. 280: Claw and Agentic Optimization
Next in Tech | Ep. 273: FinOps and AI
Next in Tech | Ep. 260: Putting Agents To Work
Next in Tech | Ep. 252: Agentic AI Use Cases
For S&P Global subscribers:
Applied AI and enterprise systems: Resiliency for the future
The CX AI Agent Index 2025
Consumers show growing interest in agentic AI but are not ready to hand over full control – Highlights from VoCUL: Connected Customer
The gap between AI optimism and return on investment in the customer experience space
Host/Author: Eric Hanselman
Guest: Sheryl Kingstone
Producer/Editor: Dylan Scheible

An enterprise's storage environments hold much more than the raw data placed in them. An understanding of the nature of the data, how it's accessed and where it's accessed from is valuable context that can be put to work to improve security and resilience. A growing number of storage infrastructure vendors are taking advantage of this potential and pivoting to a greater security focus in the way they deliver their products. Analysts Henry Baltazar and Justin Lam join host Eric Hanselman to dig into what is driving this shift and sort out the benefits and pitfalls of this change. It doesn't hurt that valuations for security companies is much greater than that of other infrastructure players, but there is a lot more at work here.
One of the biggest challenges of this pivot, is that storage and security teams, as well as their budgets, are almost always well separated. That can make realizing the benefits of enhanced security functionality complicated. As enterprises expand their operational focus into a resilience posture, there are stronger motivations to reach across these historical silos. It's a shift that can help to build more resilient digital infrastructure.
 
More S&P Global Content:
Next in Tech | Ep. 236: Data Migration
Compute sovereignty: The strategic importance of digital infrastructure
Next in Tech | Ep. 285: Resilient Digital Infrastructure
The sustainability risks threatening AI's resilience narrative
For S&P Global subscribers:
Security-related outages drive resiliency evolution
Security threats persist, recovery remains a burden – Highlights from VotE: Storage
Data Security Market Monitor & Forecast
Resiliency underpins energy transition, grid modernization and AI-driven data centers
Host/Author: Eric Hanselman
Guests: Henry Baltazar and Justin Lam
Producer/Editor: Dylan Scheible

There are a complex set of dynamics at work around data centers and the power needed to drive them. The 451 Research team has been at a pair of events, the AWS Energy Symposium and the Linux Foundation's LF Energy Summit and Jean Atelsek, Rich Karpinski, Brian Partridge and Melanie Posey join host Eric Hanselman to talk about their thoughts on the events and the state of the industry. The Symposium looked at the landscape of options for meeting power demands for the rapidly expanding data center footprint. Hyperscalers like AWS have been on the forefront of using Power Purchase Agreements (PPA's) to invest in green energy, but grid connection queues have meant that delivering that power can take considerable time.
The LF Energy Summit came at the power question as something that improved operational efficiency could address, at least in part. There are a series of projects that are looking at the first stage of putting AI to work to optimize generation and grid operation. Agentic approaches can gather operating and demand data and improve the way in which the grid responds. We're still some ways from letting agents perform any operational control, but building a foundation of greater visibility and data access could build the trust needed for next steps.
 
More S&P Global Content:
AI won't solve its own energy problem – and that might be fine
Compute sovereignty: The strategic importance of digital infrastructure
New data center legislation roundup, April 2026
The sustainability risks threatening AI's resilience narrative
For S&P Global subscribers:
AWS Energy Symposium: AI strains the grid it also seeks to fix
Linux Foundation unveils OpenGrid to accelerate power grid modernization
US data center growth: Impact on communities 101
An evolving data center industry calls for creative modifications to early-stage negotiations
Host/Author: Eric Hanselman
Guests: Jean Atelsek, Rich Karpinski, Brian Partridge, and Melanie Posey
Producer/Editor: Dylan Scheible

Data center builds are facing challenges, not only from energy supply and supply chains, but also from increasing community pushback. Analysts Tony Lenoir and Matthew Richesin join host Eric Hanselman to preview a webinar that they'll be presenting that explores wide ranging actions that being raised against data center construction and the concerns that are driving them. The traditional industry practice of "Design, announce, defend", where the build is presumed onto the community, is being reworked to include community engagement at the initial stages of a project. There are a number of regulatory strictures that are in place and more pending as a political dynamic around data centers grows in a fractious campaign season. 
All of these aspects are intertwined with societal concerns about technology and AI's impacts. As has happened with earlier technology innovations, where the ramifications of new developments aren't fully understood, there will be many who are concerned about the implications of rapid, unchecked adoption. Data centers have become emblematic of this change and a lightning rod for discussion. Join the webinar to get the full story!
 
More S&P Global Content:
Join the webinar: AI Infrastructure webinar
2026 Trends in Data Center Services & Infrastructure
New data center legislation roundup, April 2026
The sustainability risks threatening AI's resilience narrative
For S&P Global subscribers:
US data center growth: Impact on communities 101
An evolving data center industry calls for creative modifications to early-stage negotiations
Balancing data center growth and regulation: Florida's legislative conundrum
Host/Author: Eric Hanselman
Guests: Tony Lenoir and Matthew Richesin
Producer/Editor: Dylan Scheible

The data center business has undergone rapid transformation in technology and financing. The demands for capacity are generating huge capital spending and research director Kelly Morgan returns to the podcast to highlight the topics that will be on the agenda at this year's Data Center and Energy Conference with host Eric Hanselman. This year, the conference is held in Houston on October 1st and explores how the nature of the data center business has changed and how energy demands are shaping progress. Power availability has become a critical issue, with grid connection delays pushing many to on-site generation.
What's inside a data center is changing at the same time that build-out forces are at work. New GPU clusters require higher floor loading capacity. Water cooling brings greater efficiency, but requires lots of plumbing to get right and sensing and controls to manage it when it goes wrong. It's an ever-changing world with a lot of upheaval and the conference will be digging into it all. Come join us!
More S&P Global Content
Join the conference: Data Center and Energy Conference
2026 Trends in Data Center Services & Infrastructure
Compute sovereignty: The strategic importance of digital infrastructure
AI in action: unleashing agentic potential
Hyperscaler earnings quarterly: What price inference?
For S&P Global Subscribers:
Balancing data center growth and regulation: Florida's legislative conundrum
800-V DC open standard makes AI factory power infrastructure a commodity market
Service providers race to meet surging enterprise demand for AI infrastructure
Host/Author: Eric Hanselman
Guest: Kelly Morgan
Producer: Dylan Scheible