Guest post by Abhinav Gupta, Grid Transformation
Grid Forward encourages members to submit post ideas for this blog.
If I had to sum up DTECH 2026 in one sentence, it would be this: the industry has moved from asking “what technology should we implement?” to “how do we actually make all of this work together?” That second question is fundamentally harder, facing the messy reality of ensuring clean data and building trust for new tools. It was the thread that ran through nearly every session across the conference’s 14+ tracks.
The Operationalization Gap
Hundreds of millions of dollars have been spent deploying ADMS, AMI, DERMS and distribution automation across utilities of all sizes — from cooperatives like Habersham EMC serving 38,675 meters in rural Georgia to investor-owned utilities like Xcel Energy with 3.7 million customers and an $18.3 billion five-year capital plan. Yet walk into most utility control rooms today and you will find the majority of advanced distribution functions still in advisory mode. DTE Energy shared a five-year FLISR roadmap that does not reach full automation until 2029. Exelon, running ADMS across four operating companies, found that siloed testing across each OpCo creates integration bottlenecks that are painful to untangle. The hard part was never buying the technology — it is getting operators to trust it, getting the data clean enough to run it and getting the organization aligned to use what has been installed.
AI With Guardrails
AI was no longer a future-state conversation at DTech — it was a results conversation. SDG&E and Toumetis are predicting equipment failures using power quality data sampled at 61,440 samples per second, saving 2.4 SAIDI minutes in 2024 by catching incipient faults before they became outages. AEP Texas and Sensewaves are processing 105 million daily AMI data points across 1.1 million endpoints and in the process uncovered 1,480 missing and 699 incorrect GIS relationships that were silently undermining grid models. Xylem and Bidgely demonstrated natural language queries replacing SQL and GIS expertise for operators, claiming a 95% reduction in query time. Importantly, every presentation emphasized “architectures of trust” — AI that is traceable, anchored to physics-based validation and constrained to read-only access. That is exactly the right posture for critical infrastructure.
DERs and Grid-Forming Inverters
The DER conversation has moved well past visibility into active dispatch and market participation. PG&E’s DERMS now delivers value around the clock, and their Flex Connect program enables dynamic EV charging capacity through IEEE 2030.5 with approved aggregators. ERCOT’s Aggregate DER pilot has grown to 65.2 MW deployed with 124.5 MW in the pipeline. Meanwhile, EPRI, SCE and Duke Energy made the case that as synchronous generators retire, grid-forming inverters — capable of sub-5-cycle voltage and frequency support, weak grid operation and black start — will become essential to maintaining system stability. Such real-world examples give utilities and project managers the confidence to scale DERs in the market.
Cybersecurity and Data Quality
SUBNET highlighted the persistent gap between IT Security Operations Centers and OT environments, where SOC analysts have no visibility into IED event logs, firmware versions or device configurations. Tampa Electric is now penetration-testing reclosers and line sensors with the same rigor applied to IT assets. On the data side, SRP had to build custom code in Snowflake because reading their GIS topology took 36 hours. The lesson is universal: every advanced application — ADMS, DERMS, AI, digital twins — is only as good as the data and models beneath it.
What Comes Next
Layer on the workforce reality — 50% of the utility workforce is retirement-eligible within the next decade while demand is projected to grow 25% by 2030 — and the path forward becomes clear: break down the silos and build up trust. Eversource demonstrated one approach, building a “badge-less” one-team culture across 300+ people and 20+ vendors for their AMI deployment.
The next phase of grid modernization demands sustained, multi-year partnerships that span technology validation, cybersecurity lifecycle management, workforce readiness and data governance. The hardest problems are at the seams: ADMS-DERMS co-dependency, IT/OT convergence and model integrity across siloed systems. The work ahead is complex, multi-disciplinary and requires people who understand both the technology and the organizational change that has to happen in parallel.
DTECH reminds us that it’s the people we work with and trust that make it all possible.
For my full DTECH deep-dive — including DER dispatch, grid-forming inverters, some eye-opening cybersecurity gaps and a Georgia cooperative where squirrels cause 80-90% of outages — read the complete edition of Grid Talk on LinkedIn.
