A new virtual power plant would put free solar and batteries on housing its residents do not own, and keep the hardware too

Resilience Energy launched publicly in the week of 18 September 2026 with a model that is worth describing precisely, because the details are where the interest is. Hyperscalers and other large energy buyers pay for solar and battery systems to be installed on housing. Resilience aggregates those systems into a virtual power plant. The resulting capacity is sold to utilities under long-term agreements. Three parties, and the household is not one of them.

The company has raised 5 million dollars on a SAFE, has not named its investors, and is led by Ameet Konkar, formerly head of sustainability at Airbnb. It says it is in conversation with large loads, utilities and communities in the PJM and MISO regions. It has installed nothing and has no signed customers. That is the honest state of it, and everything below should be read against that.

The structural choice worth naming is ownership. Resilience owns and installs the equipment and covers the full cost; the resident pays nothing upfront and gets a lower bill and backup power. The company draws this contrast itself: aggregators such as Renew Home and Voltus orchestrate systems that households already own, and Resilience owns the systems outright, which it argues lets it build faster and reach households that would never have bought a battery. That argument is straightforwardly true. It is also the whole point.

So what moves to the edge here is the hardware. What does not move is the asset, the dispatch right, or the capacity revenue. A household ends up with generation and storage physically attached to the building it lives in and no claim on either. The company is explicitly targeting naturally occurring affordable housing and partnerships with state housing agencies and low-income housing providers, which means the model sorts by tenure. Households with capital buy their own panels and keep the upside. Households without get somebody else's panels on their roof and keep the bill saving, which is real and not nothing, while a data centre operator in another state books the capacity.

Whether that is a route into the energy transition for people otherwise locked out of it or a way of securing their roofs cheaply before they can afford to use them depends entirely on terms that have not been published. What does the resident sign, and for how long? What happens when they move out? In rented housing, is the counterparty the tenant or the landlord, and which of them is bound for the duration of a long-term capacity agreement? Who can override dispatch during an outage? What happens to the equipment at the end of the term? None of these have answers in any source located for this piece. [NEEDS DATA: resident agreement, term, termination and transfer provisions]

The scale claims need the same scepticism. The company's site claims more than 103,000 registered interests, 515 MW of potential dispatchable capacity, 50 million addressable renter-occupied homes and bill reductions of 60 to 80 per cent. Latitude Media's report gives a smaller illustration, 20,000 to 25,000 homes yielding up to 125 MW. Registering interest is not a customer, potential capacity is not capacity, and a figure computed by multiplying a wish by a kilowatt rating is a marketing number. No hyperscaler and no utility has been named by anybody. [UNVERIFIED: that any funder or offtaker is committed]

The reason to write this up at the point where nothing has been built is that the structure, not the company, is the story. Data centre operators are under real political pressure in the communities they build in, and paying for solar and batteries on local housing is a far better answer to that pressure than a philanthropic cheque. Something like this will be tried by somebody. The question this publication should keep asking as it is, in every instance, is who ends up holding the asset.