Fusion Power: How Close Are We, Really?

Fusion has passed a scientific test but not a commercial one. A US national laboratory has repeatedly produced more fusion energy than its lasers delivered to the target. Yet as of October 2026, no fusion power plant is in commercial operation. Helion says its first plant will begin initial operations in 2028. Commonwealth Fusion Systems aims for the early 2030s. Public programs point to the mid-2030s or later. All of those dates are targets.

What has fusion actually achieved so far?

Fusion forces light atomic nuclei together, usually two forms of hydrogen called deuterium and tritium, to make helium and release energy. The landmark result came on December 5, 2022, at the National Ignition Facility (NIF). It produced 3.15 megajoules of fusion output from 2.05 megajoules of laser energy delivered to its target.

Lawrence Livermore National Laboratory, which operates NIF, says the facility has now achieved ignition 11 times. The record yield is 8.6 megajoules, set on April 7, 2025, with a target gain of 4.13. That means fusion output was about four times the laser energy delivered. The most recent, on June 20, 2026, yielded 7.9 megajoules.

That gain counts only the laser energy arriving at the target. MIT Technology Review reported that the 2022 shot took about 300 megajoules from the grid, because the lasers are inefficient. NIF is a research facility. The laboratory says ignition also supports its work on the reliability of the US nuclear stockpile.

Why is net energy gain not the same as a power plant?

A power plant has to produce much more energy than the whole facility uses. It has to do so reliably and repeatedly, and convert the heat into electricity.

ITER, the international project in southern France, shows the gap. It is a tokamak, a doughnut-shaped machine that uses magnetic fields to confine hot, charged gas called plasma. ITER is designed to yield 500 megawatts of fusion power from 50 megawatts of heating power, a tenfold gain in the plasma. But it will not turn that output into electricity. Its schedule lists the start of research operation in 2034 and deuterium-tritium operation in 2039.

The Fusion Industry Association surveyed fusion companies. They named their main long-term challenges as materials that can withstand neutrons, power efficiency and tritium self-sufficiency. Tritium is radioactive, according to the International Atomic Energy Agency (IAEA). The agency says a fusion plant is expected to produce and consume it on site.

Which fusion companies are closest?

The association's 2026 report, as summarized by World Nuclear News, surveyed 56 companies. It counted $4.48 billion raised in the year to July 2026. According to that summary, this was the highest annual figure in the association's six years of reporting.

Commonwealth Fusion Systems (CFS) is building a tokamak called SPARC in Devens, Massachusetts. The company said in August 2026 that SPARC was almost 80% complete. CFS says that in 2027 the machine will produce more energy from fusion than is needed to power the process. That threshold is written as Q>1. The company's first power plant, ARC, is planned for Chesterfield County, Virginia. The goal is 400 megawatts of net electricity in the early 2030s.

Helion is building Orion in Malaga, Washington, to supply Microsoft under a 2023 power purchase agreement. Helion's website says the plant is designed to deliver at least 50 megawatts and will begin initial operations in 2028. Axios reported on September 28, 2026, that Helion had removed a website prediction that Orion would produce electricity by 2028. Axios also reported that Helion's public affairs director said full 50-megawatt operation would not come before 2029 or 2030. Helion also said in February 2026 that plasma in its Polaris prototype had reached 150 million degrees Celsius using deuterium-tritium fuel, TechCrunch reported.

These are company statements. SPARC was still unfinished when CFS gave its update, and a temperature milestone is not a net energy result.

When will fusion energy be available?

No firm date exists. Estimates fall into three groups:

1. Companies. The Helion and CFS targets appear above. In the industry association's survey, 71% of companies expected a fusion plant to be delivering power to the grid by the 2030s.
2. The US government. The Department of Energy's fusion roadmap, finalized in June 2026, aims to support pilot plants and commercial fusion power in the mid-2030s. The department says that depends on future congressional appropriations and public-private partnerships.
3. International public programs. The IAEA's fusion FAQ says a prototype reactor, known as DEMO, is expected by 2040, with electricity generation in the second half of the century. It says that depends on funding and technical progress.

Rules are unfinished too. The US Nuclear Regulatory Commission published a proposed rule on requirements for fusion machines on February 26, 2026. Its rulemaking page, last updated on September 14, 2026, lists no final rule.

The bottom line

Ignition is a real scientific milestone, but it is measured at the target, not at the plug. Two tests come next. One is whether SPARC demonstrates Q>1, which CFS targets for 2027. The other is what Orion delivers when Helion begins initial operations, scheduled for 2028. Until a machine sends out more electricity than it takes in, commercial fusion remains unproven.