Why the energy industry must fast-track standards, data access and collaboration. What our own children can teach us by watching them for a few hours.
For more than three decades, I have argued that one of the energy industry’s biggest challenges is not a shortage of ideas, technologies, or intelligent people. It is something more fundamental: getting usable energy data to the people and systems that can do something with it.
The arrival of artificial intelligence makes that challenge considerably more urgent. We can invest in AI platforms, digital twins, Distributed Energy Resources, Virtual Power Plants, building optimization, carbon accounting and advanced energy-management applications. But all of them depend on trusted, accessible data that creates an effective industry for everyone.
When every application must separately find, extract, clean, interpret, validate and integrate that data, we repeatedly pay to solve the same problem before addressing the problem that matters.
The better sequence is straightforward: standardize the data, make it securely accessible, and let innovation compete on what it can do with the data.
From discussion to practical action through the minds of our children
Our collaboration with the Caribbean UEnergyHub on the AI Energy and Climate Hackathons, first in Trinidad and Tobago and now jointly with Canada, shows this approach in action within 8 months.
At the first CZITT–UNICEF AI Energy and Climate Hackathon, held at the University of Trinidad and Tobago’s Point Lisas campus, students worked with energy information from their own schools. Before the competition, they received mentorship covering AI and coding, electricity, climate change, data analysis, leadership and youth climate action. In June, 2026, they then used real information to investigate how their schools consumed energy and to develop evidence-based recommendations.
This was not a theoretical sustainability exercise. Students connected consumption patterns, peaks and anomalies to real school operations such as air conditioning, lighting and they did this with only a little guidance. That is where an energy dataset stops being an IT asset and starts becoming actionable infrastructure.
Hackathon II expands the experiment
The second CZITT–UNICEF AI Energy and Climate Hackathon on November 27th extends the model across borders, connecting schools in Trinidad and Tobago with schools in Newfoundland and Labrador, Canada. Students analyze electricity consumption and weather information to build AI-supported models and practical carbon-reduction proposals.
Underneath the competition is an equally important experiment: can participants in different jurisdictions investigate comparable energy questions using common data structures and a shared approach to secure access? They should not have to become experts in proprietary utility formats before they can begin solving energy and climate problems. That is what a practical standard is meant to accomplish.
“We can no longer afford to spend years agreeing on how energy data should move while AI and the digital economy advance around us. Utilities, governments, regulators, standards bodies and technology companies need to work together, fast-track practical standards, implement them and improve them through real-world use. The energy industry must adopt faster so the new economy does not have to wait for us to catch up.” Gary Michor, CEO, Screaming Power Inc.
We no longer have years to agree
The energy industry has historically spent years discussing interoperability, debating standards, building consensus and eventually moving toward implementation. That approach may have been acceptable when technology and markets changed more slowly. It is incompatible with the pace of AI, electrification and the new digital economy.
AI is moving in months, not decades. New applications are being developed continuously. Buildings, distributed energy resources, electric vehicles, storage, utilities and customers are becoming increasingly interconnected. Yet much of the information needed to make these systems intelligent remains trapped behind inconsistent formats, interfaces, proprietary access and organizational boundaries.
No single utility, technology company, regulator, standards organization, university or government can solve this alone. Rapid change requires collaboration across the ecosystem. Parties need to identify what is already working, agree on the minimum standards required for interoperability, test those standards in real operating environments and move quickly into adoption.
Fast-tracking does not mean abandoning governance, cybersecurity, privacy, consent or technical scrutiny. It means replacing a “perfect, then implement” mindset with a more practical cycle: collaborate, implement, learn and improve.
The Student AI Hackathons demonstrate this philosophy on a manageable scale where the first Hackathon used a CSV file and the November Hackathon will use and open API. Â It is enlightening to see organizations from education, government, technology, energy and the nonprofit sector coming together in a matter of weeks around a real use case. We are watching students gain access to actual information while the underlying data approach is tested through implementation rather than discussed only in theory.
The real savings happen downstream
An energy-data standard can sound like a narrow IT initiative. Its economic significance is much broader.
Without standardization, an energy-management company develops one utility integration. An ESG platform develops another. A researcher devises a separate process. A building owner hires consultants to assemble spreadsheets. A DER provider builds another connector. An AI company cleans yet another dataset. Each organization spends money overcoming essentially the same barrier.
Screaming Power’s work with data hubs, standards groups, open APIs, automated testing, data validation and operational data stores has shown how a common data layer can reduce repeated manual integration and testing while allowing legacy utility systems to remain in place. The point is not that one platform or standard eliminates every cost. It is that solving data access once prevents many organizations from paying to solve substantially the same problem again.
A dollar invested in shared data infrastructure creates benefits across conservation, billing analysis, building benchmarking, carbon reporting, DER integration, grid flexibility, research and AI. That is the multiplier effect.
Green Button showed part of the path
Ontario’s Green Button rollout offers a useful lesson in partial success. The province aimed to bring more than 55 electricity and natural gas utilities together to provide automated, authorized access to energy and billing data, and best practices were published (see www.greenbuttonontario.ca). Yet no one ever tested whether those implementations and best practices functioned together as one standardized data connection. I know this firsthand: my company advised the Ministry of Energy in 2018 while the regulation was being developed, and I co-chaired the OEB GB Working Group for over two years. Thousands of hours went into the effort, some ( like myself) volunteered and some paid for by ratepayers on behalf of participating utilities. Transitions like this have always moved slowly, mostly because of red tape. In reality, every implementation requires clear use cases and testing after launch, and governments must accept that regulation by itself does not ensure success. We need a new approach, because saving money and energy benefits everyone. And this work enables far more than a chart of a customer’s consumption, as the students in the first hackathon demonstrated.
Banking and telecommunications depend on standards that most people never notice. No one thinks about the infrastructure behind an ATM withdrawal or mobile roaming; we simply expect it to work. Energy data should work the same way, and arguably better, because it feeds so many upstream and downstream processes across today’s energy ecosystem.
Newer initiatives such as LF Energy’s Connected Data Specifications can extend this connectivity faster than ever. By enabling secure, standardized machine-to-machine communication across modern energy systems, the goal should be portable, validated, AI-ready information, so applications compete on the insight they deliver rather than on their ability to decode proprietary formats. Hackathon II will use this new specification, so let’s see what high school students can do with it. I expect they will show us experts that data access can be simple, and that once trusted, source-of-truth data is securely available to those who want to help, the barriers to innovation fall away.
Make standards an immediate action item
Energy policy discussions often place standards near the end. We discuss AI strategies, smart grids, electrification, DERs, VPPs, digital twins and decarbonization. Only later do we ask how the systems will exchange information. We need to reverse that order.
Governments, regulators, utilities and standards organizations should treat secure, interoperable energy information as a short-term implementation priority. The agenda should include open and practical standards for billing and consumption data; secure authorization and consent; machine-readable APIs; data-quality validation; portable AI-ready information; and jurisdictional data sovereignty.
Interoperability does not require surrendering sovereignty; in fact it ensures it is in place. Agreeing on how information communicates while preserving who controls it, where it is stored and how consent is managed.
The students may be showing us the answer
The AI Hackathon model strips away some of the complexity that has accumulated around energy data for 100 years. Give students electricity and weather data. Teach them how to understand it ina few 30 minute sessions. Give them modern AI tools. Then ask a direct question: What can you learn, and what would you change?
The first Hackathon showed that students could move from passive energy consumers to analysts making evidence-backed recommendations in hours. Hackathon II adds students in two countries working on the same fundamental problem through common data structures and technology.
The lesson is not that every industry problem needs a hackathon. It is that when information becomes accessible, secure, understandable and reusable, the number of people capable of creating value from it expands dramatically. Researchers can research. AI developers can develop. Energy managers can manage. Building owners can find savings. DER providers can integrate. Students can innovate. Utilities do not need to build every downstream solution themselves.
A call to move now
Technology is moving. The new economy is moving. Our standards and energy-data infrastructure must move with them.
We have spent enough time proving that energy data has value. We now need utilities, governments, regulators, standards organizations, researchers, educators and technology companies working together to unlock that value. We should not spend another five years designing a perfect standard before implementation begins because by that timer the design itself has no value because the world has moved on. We should establish the common foundation, try it in real use cases and deploy it in controlled real-world environments, learn from implementation, improve it and scale what works.
That is one of the lessons from our work in Trinidad and Tobago and now between Canada and the Caribbean. The Hackathons involve students and schools to establish the workers of our future, but the principle is much larger: when trusted data can move securely through a common standard, innovation can move with it and both can change as required without bureaucracy and protectionism.
Build the standards. Open secure access to the data. Implement them in the real world. Improve them quickly. And let innovation happen.
Every year spent debating how data should move is another year that AI developers, researchers, building owners, customers and emerging energy businesses spend money solving a problem we should have solved together in days rather than years. If we get the foundation right, the greatest return may not come from the first application we build. It may come from all the applications that no longer need to rebuild the data infrastructure from scratch.