Product1 distinct publisher3 min readUpdated
EarthGrid melted its way three metres into a California quarry in January and now talks about commercial work next year. The undergrounding demand is real; the delivery evidence is thin.
The Product Desk · Product desk

Compiled by The Product DeskSomething wrong?How this is made
In a California quarry in January, EarthGrid's cigar-shaped plasma boring machine cut three metres into white granite [1], and founder and chief executive Troy Helming told the BBC his team hopes the machine could reach commercial deployment as early as next year [2][3]. That gap between a three-metre demonstration and a priced route is the whole story, because burying power and telecom lines has always been the expensive option, and demand for it is now rising for reasons that have nothing to do with drilling technique [9][10].
The mechanism is not subtle. Three plasma torches sit in a spinning head and produce super-heated plasma reaching 27,000C, hotter than the surface of the Sun [4]. Helming describes ignition as "kind of like igniting a rocket" and initially loud [5], and says the machine creates a violent vortex inside the tunnel that helps suck away debris, which at times takes the form of lava [6]. He also says the January run over-bored slightly to the top and left, and that the fix is to make the vortex reverse direction every five minutes or so [7]. Helming had waited ten years for the test [8].
What the account does not contain is the operator-relevant part: no elapsed time for the three metres, no advance rate, no cost per metre [2]. Three metres is also, by coincidence, the depth to which submarine cables are now routinely buried [1][17], which is a useful sense of scale for anyone imagining a transmission corridor. Helming says he has had interest from companies wanting the machine for power and fibre optic cables, or pipelines carrying water, natural gas or ammonia [11], and the company has eyed a tunnelled freight distribution system around airports and warehouses to take trucks off the road [12]. Those are markets, not contracts.
Meanwhile the undergrounding trend is proceeding without any new drilling method. Engineering firms told BBC News they are seeing rising demand for putting infrastructure below ground, partly because Russia's war with Ukraine has shown how exposed above-ground facilities are to drone attacks [10]. Alexander RE Taylor of the University of Exeter calls the shift a "data bunker boom" and argues that "brutal materiality is still important" [13]. A data centre completed earlier this year sits in caverns 100m under the Dolomite Mountains in Italy, beside stores that have recently held sparkling wine, apples and cheese [14]; the cool rock cuts the cost and energy of cooling servers [15], and Trentino DataMine chief executive Dennis Bonn says the surrounding 90 million cubic metres of dolomite give protection against intrusion, electromagnetic interference, seismic events and hydrogeological risk that cannot be replicated above ground [16]. At sea, Lane Burdette of TeleGeography says faults per kilometre of deployed cable have become rarer as burial has become more common, with entire lengths buried in some fault-prone areas [17][18].
Two things would move plasma boring from demonstration to procurement. First, a metered figure: metres per shift in hard rock, with the corrected counter-rotating vortex actually holding the bore straight [7]. Second, a named customer and a stated route length behind the next-year deployment claim [3]. Until then, the cheap way to get underground remains an excavated cavern or a plough on a cable ship [14][17].
Ranked by verification strength, evidence, and original report placement.
Engineering firms told BBC News they are seeing rising demand for undergrounding, in part due to Russia's war with Ukraine, which has revealed how vulnerable above-ground facilities can be to drone attacks.
During a test in a California quarry in January, EarthGrid's cigar-shaped rock-melting tunnel boring machine chewed through three metres of granite.
Three plasma torches set within a spinning head at the front of the machine produce a stream of super-heated plasma reaching 27,000C, significantly hotter than the surface of the Sun.
Helming says lighting up the three plasma torches is "kind of like igniting a rocket" and is initially loud, with the torches quietening down afterwards.
Helming says the machine creates "basically, a tornado - a violent vortex inside the tunnel" that helps suck away debris, which at times takes the form of lava.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 18, 2026
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One quarry bore, one publisher, no rates
The technology claim rests on a single BBC feature describing a single January test whose only quantitative output is three metres of granite. Physical detail (torch count, 27,000C plasma, vortex debris removal, the over-bore defect) is specific and internally consistent, but there is no elapsed time, advance rate, cost per metre, energy figure or independent verification, and no second publisher. The surrounding undergrounding evidence is stronger because it is attributed to multiple named practitioners and completed projects.
Plasma at demo stage; undergrounding already shipping
EarthGrid has one quarry test, unnamed inbound interest and an eyed freight-tunnel concept - no commercial deployment, contract or customer. Adoption that is real in this cluster belongs to conventional undergrounding: a completed 100m-deep Dolomites data centre, 29km of London Power Tunnels, and routine subsea cable burial to three metres. The composite score reflects a story whose headline technology has zero adoption while its market thesis has demonstrable deployments.
Commercial-next-year talk outruns three metres
The gap is moderate and one-directional: a founder-stated hope of commercial deployment as early as next year sits on top of a three-metre bore with a known steering defect and no published rate or cost, in a sector where a working tunnelling contractor in the same article says step-change boring innovation is rare and progress is measured in millimetres per minute. The article itself supplies the deflating context and independently sourced counterpoints, which keeps the gap from being extreme.
Founder and vendor voices carry the story
The performance narrative comes almost entirely from EarthGrid's founder and CEO, who is promoting a pre-commercial machine and a 10-year personal investment in it. The corroborating undergrounding demand signal comes from parties who sell undergrounding or underground capacity: a tunnelling contractor's head of work winning and business development, an underground data centre's chief executive, and a telecoms market research analyst. An academic and an engineering consultant who says the UK shows no shift yet provide the only commercially disinterested or dissenting input.
Clear reporting, single source, unmeasured core
The reporting is specific and attribution is clean, so the descriptive facts are reliable. Confidence is held down because there is only one publisher, the central technology claim has no measured throughput or independent verification, and the forward-looking commercial statement is unfalsifiable from the material supplied.
Follow any of these and your For You feed starts watching them — no settings page required.
science
When the method is a purchase: BioScience team says tool opacity needs an audit trail1 distinct publisher