If your business had a very expensive machine sitting underutilized 99% of the time, you’d probably put a lot of brainpower into increasing its use.

An initiative called Grid 2.0, launched in a new research paper, aims to address this problem for the U.S. power industry. Utilities and transmission operators build capacity for peak energy use, which means those resources sit idle except for the 30-some hours a year of peak demand.

Flexibility to tap into that underused capacity needs to be huge factor in the future for data center growth. A raft of research shows that big power users can unlock lots of electric grid capacity if they’re flexible around peak demand. Grid 2.0 offers a protocol and operating model to deliver flexibility.

“It’s trying to implement in a standardized way what everyone is asking for,” says Chris Shelton, senior VP and Chief Product Officer of the utility AES and a Grid 2.0 leader, on Jigar Shah’s Energy Empire podcast.

Here are four key ideas about the project, including how you can get involved.

The protocol’s modeled on the internet: Grid 2.0 is a protocol standard for orchestrating and automating the management of large electricity loads, transmission systems, batteries, and clean energy assets. Participants would use the voluntary protocol to ask the grid, minute by minute, whether a workload can run or an energy resource is needed. It would manage grid resources using scheduling techniques inspired by internet protocols such as TCP/IP[CG1] , and cloud-computing resource management capabilities such as Kubernetes[CG2] . Another similarity: the creators formed a Grid 2.0 Task Force to emulate the Internet Engineering Task Force that figured out how to make the internet work. The idea is that no single organization is going to solve a problem this big.

It fits into the existing electrical grid system: The protocol’s described as a “digital overlay” that makes allocation decisions every minute — sitting between the current grid’s split-second capacity reactions and its five-minute broader market actions. Because Grid 2.0 works within the existing grid framework, it can be adopted voluntarily by only a few energy users or providers to start. “A key design feature of Grid 2 is that it does not need to be universally applied,” the paper says. “It serves and sits inside the Grid 1 context at any scale or scope.”

It’s possible now because of new technology: It’s not like we just learned about the limitations of a system overbuilt to protect for peak demand. What’s changed is we now have increasingly flexible energy demand and sources, including massive battery capacity connected to the grid, large-scale renewable energy generation, and huge workloads from sources such as data centers managed with sophisticated scheduling software. These provide need for flexibility and also the capabilities to deliver it on a meaningful scale. Grid 2.0 is a proposal to create one framework that can work across organizations to orchestrate that flexibility.

Grid 2.0 doesn’t have all the answers: The paper even includes a section called “Open Questions,” listing 16 of them under six categories, describing it as “an initial list.” There are some doozies on the list, such as “What processes are necessary to ensure utilization benefits translate to reduced customer rates while improving utility financial performance?”.

The Grid 2.0 team and paper present the protocol as a starting point. The report offers a confident view that it’s the right starting point, with the right goals and direction, but the authors also fully expect it to evolve and change with input from across the industry.

“We have to work together as an industry and find this paradigm,” Shelton said on Energy Empire. “Whatever version of Grid 2 we’re talking about in the future, that we’ve modified and made work, it’s worth all of us working on it, because the opportunity is enormous.”