In 2023, Liz and Rich Muller founded Deep Fission to put working nuclear reactors underground. 4 years later, its Kansas pilot is targeting a 5-MW reactor by late 2027

A striking new idea in the race for next-generation nuclear energy is taking shape beneath the ground in Parsons, Kansas. Deep Fission, founded in 2023 by father-and-daughter team Rich and Liz Muller, is developing a concept that places a small pressurized-water nuclear reactor deep inside a borehole rather than housing it in a conventional surface nuclear plant. The company’s current plan is to demonstrate its technology through the U.S. Department of Energy’s Reactor Pilot Program, with a commercial pilot targeted for 2027. The project represents a significant departure from traditional nuclear construction and comes as electricity demand from data centers, industry and artificial intelligence continues to increase.

The basic concept sounds almost futuristic: put the reactor roughly a mile underground and use the surrounding geology as part of the safety system. Instead of constructing a massive above-ground containment structure, Deep Fission’s Gravity Reactor is designed to take advantage of the rock and water surrounding the underground installation. The company says its approach is intended to provide passive shielding and containment while reducing the amount of expensive infrastructure required at the surface. The idea grew from the Mullers’ earlier work on Deep Isolation, which focused on using deep boreholes for nuclear-waste disposal. Deep Fission applies related borehole technology to a very different objective—generating electricity.

Kansas Project


The Kansas project is not simply a matter of drilling a hole and lowering a reactor into it. The borehole itself is one of the biggest engineering challenges. Deep Fission has already drilled a data-acquisition well to approximately 6,000 feet to collect geological, hydrological and thermal information. The company’s next phase involves demonstrating that it can construct a commercial-scale borehole and safely deploy a prototype system. Its current Parsons site information says a roughly 2,500-foot non-nuclear proof-of-concept borehole is being targeted for drilling in the third quarter of 2026. That test is intended to examine thermal behavior and demonstrate deep deployment of major components before a nuclear system is installed.

The 5-megawatt target for 2027 is an important distinction in Deep Fission’s development strategy. The company has previously discussed a larger 15-megawatt version, but the initial pilot is expected to operate at a smaller output. Deep Fission’s broader commercialization roadmap identifies 2027–2028 as the period for pilot construction, authorization and licensing activities. The company says the pilot reactor will use established pressurized-water-reactor technology, while much of its development effort is focused on proving the unusual underground deployment system. In other words, the central technological question is not whether a conventional reactor can generate heat, but whether the entire nuclear system can be safely drilled, installed, monitored and operated deep underground.

Electricity Demad Rises in U.S.A

One reason the project has attracted attention is its potential relationship with America’s rapidly growing electricity demand. Data centers and artificial-intelligence infrastructure require reliable electricity around the clock, while industrial customers are also seeking dependable low-carbon power. Deep Fission says its commercial strategy is designed around supplying reliable baseload electricity to utilities, industrial customers and data centers. The company has announced letters of intent representing 18.5 gigawatts of potential demand, although those agreements are non-binding and should not be interpreted as guaranteed future electricity sales. The interest nevertheless illustrates why smaller nuclear technologies are attracting attention as developers search for dependable power sources.

The underground approach could offer another potential advantage: a smaller surface footprint. Conventional nuclear facilities require extensive buildings, cooling systems, security infrastructure and other equipment. Deep Fission envisions a surface installation more closely resembling an industrial or geothermal-style energy facility, with the reactor itself positioned far below ground. The company argues that using geology for shielding and containment could reduce some of the physical infrastructure associated with traditional nuclear plants. But those potential advantages remain claims to be demonstrated. Deep Fission itself acknowledges that its pilot must validate drilling, deployment, integration and operation before the technology can move toward widespread commercial use.

Underground Nuclear Reactor

Safety and regulation will therefore remain central to the project’s future. A nuclear reactor underground is not automatically a proven safer reactor; the complete system must demonstrate that it can meet regulatory requirements under real operating conditions. Deep Fission is participating in the DOE Reactor Pilot Program and says it is working with the Department of Energy and the Nuclear Regulatory Commission on its development and licensing pathway. The company plans to seek a commercial license while advancing the pilot, subject to regulatory approvals, engineering completion and available funding. Its filings emphasize that the timing of DOE authorization and testing remains dependent on those factors.

Kansas experiment represents something larger than one company’s attempt to reinvent the nuclear power plant. Deep Fission is testing whether decades-old reactor technology can be paired with modern drilling techniques to create a radically different energy infrastructure. If the company succeeds, underground small reactors could offer a new way to produce dependable electricity close to major industrial and computing loads while potentially reducing the surface infrastructure traditionally associated with nuclear generation. If the concept encounters unexpected geological, engineering, regulatory or economic obstacles, those lessons could be equally valuable. For now, the proposed 5-MW 2027 pilot remains a target rather than a completed achievement—but the work underway in Kansas could provide an important test of how far the next generation of nuclear energy can go beneath our feet.