In about eight years The Boring Company went from a tweet about traffic to a working underground network in Las Vegas — nine stations and roughly 3.5 million rides — while claiming it can tunnel for ~$10M per mile against the $200M–$1B per mile a US subway costs[8][11][14].

The genuinely open question is the one the company has carried since day one: is this a real cost-and-speed breakthrough in tunnelling, or — as transit experts and the tunnelling press argue — a low-capacity demo dressed up as infrastructure? The evidence cuts both ways. TBC has built more deployed mobility than any rival tunnel venture and just broke ground on a privately-funded Nashville line; it has also cancelled project after project, never delivered the promised autonomy, and run into a string of safety and environmental citations. This study lays out both cases on each question; the verdict is yours.

The decisive questions

Each links to the section that lays out the evidence on both sides.

The mark that frames the debate

Reported private valuation marks ($B). The climb to $8.6B in 2023 and the markdown to ~$5.7B in 2025 capture the swing in how investors price the same unproven thesis. Hover any point.

TBC is unusual among Musk ventures: privately held, fewer than ~400 employees as of 2022, and run day-to-day by longtime lieutenant Steve Davis — who also became Musk's de facto No. 2 at DOGE in 2025[1][3]. It has shipped one operating system (Vegas) and a graveyard of announced-but-cancelled ones.

How it started

The company was announced in December 2016 and incorporated on January 11, 2017, headquartered today in Bastrop, Texas, with fewer than roughly 400 employees as of April 2022[1]. Its first physical output was a 1.14-mile test tunnel in Hawthorne, California (2018), funded in part by selling 20,000 "Not-a-Flamethrower" novelty devices — an early signal of how much the venture ran on Musk's marketing reach rather than infrastructure revenue[2].

The people behind it

Since November 2019 the company has been run by president Steve Davis, an aerospace engineer who spent years at SpaceX before joining TBC in 2016, reorganized Twitter/X for Musk after the 2022 acquisition, and in 2025 became Musk's de facto second-in-command at DOGE before departing alongside Musk when their special-government-employee terms expired[3][34]. That overlap is part of why TBC is best understood as one node in the broader "Elon empire" rather than a standalone infrastructure firm.

Timeline

US subway tunnels run roughly $200M to $1B per mile; TBC says its narrower, single-bore design tunnels at about $10M/mile today with a goal of $3–4M[5][14]. If even partly true, that reshapes which tunnels are economically worth building — if the cheap figures hold outside a subsidized demo.

Why US tunnels are so expensive

Estimates for conventional US subway construction range from about $200 million to $500 million per mile, and sometimes up to $1 billion per mile — far above European and Japanese comparables[5][35]. Drivers include large twin bores sized for trains and platforms, extensive stations, labor and procurement rules, and the stop-and-go nature of standard tunnel boring machines (TBMs), which dig a few feet, halt to erect precast concrete lining segments, then resume[38].

The Boring Company's wedge

TBC's pitch is to attack that cost base structurally: a narrow single bore just wide enough for a car-sized vehicle, a machine designed to install the lining while it mines rather than stopping, and surface-launch reuse so it avoids excavating expensive pits[38][15]. The company targets an all-in Loop tunnel cost of less than $8M/mile, and president Steve Davis has put current Vegas tunnelling at about $10M/mile with a stated goal of $3–4M within two-to-five years[4][14].

TBC figures are company targets and a president's quote (highlighted); subway range is a reported industry estimate. The comparison is not perfectly like-for-like — a car-width Loop tunnel carries far fewer people than a train tunnel (see Vegas Loop), so a lower cost per mile does not automatically mean a lower cost per rider.

TBC's stated model spans selling/leasing Prufrock machines, turnkey tunnel construction, operating transit (the Vegas Loop), and licensing software/controls[6]. Today the operating transit line is the only one clearly producing revenue — and the public LVCVA contract pays TBC about $4.5M a year (≈$7.50/ride) on top of any fares[7].

The four revenue lines

• Prufrock sales / leasing. Selling or leasing its boring machines to other operators — scalable, but no major external customer is publicly confirmed[6].
• Turnkey tunnel construction. Building tunnels for cities and private clients (the Vegas and Nashville model)[6].
• Passenger transport. Operating the Loop service itself — the LVCC system, resort connectors and (soon) airport links[6][8].
• Software / control licensing. Licensing the system that routes and dispatches the vehicles[6].

How the Vegas economics actually work

The Las Vegas model is a hybrid. Rides between the convention-center stations are free to attendees, with the Las Vegas Convention and Visitors Authority paying TBC an annual subsidy of about $4.5 million — roughly $7.50 per ride[7]. At resort-connected stations such as Resorts World, riders pay directly, with day passes around $5[7]. In August 2025 TBC also sought approval to run its tunnel Teslas above ground as surface taxis, and began limited above-ground airport rides in December 2025 — a notable step away from the pure-tunnel premise[36].

The LVCC Loop opened June 1, 2021 and had grown to nine stations by February 2026, carrying about 3.5 million riders since launch[8][11]. But its measured peak was 1,355 passengers/hour (July 2021) against a full-build claim of ~90,000/hour — and the cars are still driven by humans, not the promised autonomy[10][20].

What is actually running

The first segment — 1.7 miles, three stations, built for $53 million — used Tesla Model Ys at 35 mph[8]. As of early 2026 there are nine operating stations spanning the convention center and connected resorts (Resorts World, Westgate, Encore, Fontainebleau), with over 10 miles dug and roughly 4 miles operational[8][11]. Ridership has compounded from 1 million (March 2023) to 2 million (May 2024) to about 3.5 million by early 2026[12][11].

The throughput question

This is the crux of the "disruption vs vaporware" debate. TBC says the full 68-mile, 104-station build will move up to ~90,000 passengers per hour, and president Steve Davis says the airport line will lift the system to 17,000–20,000/hour[9][13]. The strongest demonstrated figures are far lower: a test demo of about 4,400/hour, a measured July-2021 peak of 1,355/hour, and an event high of about 30,000 riders in a day (and ~82,000 over the CONEXPO show) across multiple stations[10][39].

The 1,355/hour measured peak (highlighted) is the only audited line on this chart; the larger figures are company claims, and the 90,000 full-build number has been flagged by others as potentially inflated[9][10].

The autonomy gap

The Loop was promoted as a fleet of autonomous vehicles, but as of 2025 every Tesla still has a human driver: Clark County approved only human-driven cars, set strict speed limits, and forbade onboard collision-avoidance, all of which cap throughput[20]. Removing drivers is both the key to the cost model and the thing TBC has not yet delivered — a recurring theme across Musk ventures.

Prufrock's design ideas depart from conventional TBMs: it "porpoises" — launching from the surface in roughly a day with no pit or crane — and is built to line the tunnel while it mines rather than stopping[15]. The speed and cost figures (>1 mile/week now, 7 miles/day eventually, <$8M/mile) are company targets, and tunnelling veterans caution that a fast machine is not the same as a delivered, regulator-approved tunnel[14][37].

What Prufrock does differently

Standard TBMs work in a stop-and-go cycle: dig a short distance, halt to erect precast concrete lining segments, then resume — slow and labor-intensive[38]. TBC's two core goals, per Contrary Research, are to increase boring-machine speed and power and to line the tunnel simultaneously with digging[16]. Prufrock also uses "porpoising": it arrives on a truck, tilts down, and launches into the ground within about a day, then resurfaces onto a trailer for reuse — eliminating the expensive pits and cranes conventional projects need[15]. The company targets "Zero-People-In-Tunnel" operation[15].

The numbers — and the caveat

TBC targets a dig rate greater than 1 mile per week (about 6x its previous-generation machine), and says future generations aim for roughly 7 miles per day[14]. Current Vegas tunnelling runs about $10M/mile, with a stated goal of $3–4M within two-to-five years using a later Prufrock generation[14]. The important caveat: these are forward targets. Industry commentary contrasts Prufrock's speed focus with incumbent Herrenknecht's proven precision and reliability, cautioning that "tunnelling isn't always a race" — raw dig-rate goals don't equal lined, inspected, code-compliant tunnel delivered at scale[37].

No one else runs a sub-$10M/mile, single-bore tunnel with an integrated EV shuttle — so direct rivalry is low. But the market is full of substitutes (rail, BRT, monorail, ride-hail), and transit experts argue those move far more people per hour, making this a substitute-rich market, not an open field[17].

Porter's Five Forces

Rated from the industry/incumbent lens. Click a force to see the evidence behind its rating.

The expert critique

Public-transit consultant Jarrett Walker rode the Vegas Loop in January 2025 and was unconvinced. He framed it as "a very low-capacity solution to a problem that required much higher capacity," arguing the car-width tunnels physically cannot scale or be converted to rail[17]. He observed single-track bottlenecks on his own ride and warned that a system that "sort-of worked with four stations would be a fiasco with more than 60"[18]. Earlier, the Tunnelling Journal dismissed TBC as a "vanity project," and Walker has called it "wildly hyped" — saying it "dazzled city governments… but turned out to be a paved road tunnel"[19].

Where the Loop sits among the alternatives

Two axes that actually differentiate this market: how proven/operational the system is, and person-throughput per corridor. Hover a point for the sourced basis.

The Loop is more operational than other new-mobility tunnel concepts (the Hyperloop was abandoned), but sits low on throughput versus the rail and bus modes transit experts say the corridor needed[17][26].

On the build side: the ~10-mile Music City Loop between downtown Nashville and the airport is 100% privately funded, won a 40-year airport license, and began construction on Feb 25, 2026 with Prufrock already in the ground[21][22]. On the cancellation side: Chicago, DC–Baltimore, Ontario, Fort Lauderdale and LA were all announced and then dropped[25].

The Nashville Music City Loop

Announced July 28, 2025 with Tennessee Gov. Bill Lee, the Music City Loop is a roughly 10-mile, 100%-privately-funded tunnel between downtown Nashville, the convention center and the airport, with a claimed ~8-minute transit time and 20+ stations in design[21]. The Metro Nashville Airport Authority approved a 40-year license (extendable to 50) over ~933,000 sq ft at $300,000/year (rising 3% annually, ~$34M over the term), and Gov. Lee announced state right-of-way approval on Feb 25, 2026 — with construction beginning the same day and Prufrock-MB1 already in the ground[22]. This is the clearest evidence to date that TBC can win a second-city deal and move dirt on it.

But the council pushed back

The project is contested locally. On March 4, 2026 the Metro Nashville Council voted 20–15 (two abstentions) to formally oppose the Music City Loop, citing worker-safety history, transparency and the potential taxpayer cost of an oversight commission[23]. The resolution is non-binding, and because the state approved the project, construction continues regardless[23]. The council's split mirrored the national one: one member cited TBC's labor record, while another who visited the site said she was satisfied with safety[24].

The cancellation graveyard

The skeptics' core exhibit is the list of projects TBC announced and never built — the pattern critics point to when they call it "transportation vaporware"[25][26]:

The original Hyperloop ambition — the vacuum-tube vision that generated much of the early hype — was effectively abandoned: the DC–Baltimore Hyperloop was dropped, and the Hawthorne test track was torn down in 2022 to make room for a SpaceX parking lot[26].

TBC raised a $675.9M Series C (April 2022, Sequoia & Vy Capital) at a $5.675B valuation; total funding is about $905.6M[27]. Its reported mark climbed to $8.6B in July 2023 and then fell back to about $5.70B by 2025 — a markdown, not a closed round[27].

Funding and valuation

The company is private and discloses no revenue or profitability[28]. Its backers are a roster of venture and institutional investors — Sequoia, Vy Capital, Valor Equity Partners, Founders Fund, 8VC, Craft Ventures, DFJ Growth, Brookfield, Lennar and Tishman Speyer[28]. The valuation arc tells the story of sentiment: ~$920M after the 2019 Series A, $5.675B after the 2022 Series C, a peak $8.6B mark in mid-2023, and a step down to ~$5.70B by 2025[27].

These are reported private marks (the 2025 figure per Sacra), not transaction-confirmed values at each point. Private valuations can lag operating reality in both directions.

What the markdown does and doesn't tell us

The drop from $8.6B to $5.7B coincided with a broad repricing of speculative tech and with TBC's mixed execution (cancellations, undelivered autonomy). But because the underlying financials are undisclosed, the mark is a sentiment proxy, not a verdict on revenue. The most concrete economic facts remain the Vegas subsidy (~$4.5M/yr from LVCVA) and the Nashville license (~$34M over the term) — small numbers relative to the multi-billion valuation, underscoring how much of the value is still about the future cost curve[7][22][27].

TBC's edge is cost per mile and the fact that it actually operates a system; its disadvantage is throughput per corridor and the gap between claimed and demonstrated capacity[14][10]. The right denominator — cost per rider — is exactly what remains unproven.

The comparison table

Cost and throughput figures are mixed (company targets, measured peaks, and industry estimates) and not directly comparable cell-to-cell — see Market and Vegas Loop for the underlying numbers and their caveats[5][10][14].

SWOT

A specific, sourced synthesis — weaknesses and threats given equal weight to strengths.

Each SWOT item is sourced in the section it draws from; see Vegas Loop, Technology, Expansion, Financials and Risks.

TBC faced chemical-burn citations, a contested ~$425K OSHA rescission involving the Nevada governor's office, a separate $112K set of 2024 violations, and a ~$500K sewer-dumping fine — while its president ran DOGE, binding the company to Musk's politics[29][30][32][33][34]. These are the risks that don't depend on the capacity debate at all.

The firefighter chemical-burn case

In December 2024, two Clark County firefighters suffered chemical burns (later found to be permanent scarring) during a training drill in the tunnels, from an accelerant (MasterRoc AGA 41S) that pools at the tunnel base; a Fortune investigation documented numerous prior worker burns from the same chemical[29]. Nevada OSHA issued three willful citations totaling about $425,595 in May 2025 — then withdrew them after Boring president Steve Davis called Gov. Joe Lombardo's office and state officials met TBC the next day. Fortune found that a case-diary line documenting the governor's-office meeting had been deleted[30].

Federal OSHA later concluded Nevada had "reasonable justification" to withdraw the citations — but confirmed that records had been "altered, missing and/or removed"[31]. TBC blamed the Clark County Fire Department for the training-drill breakdown; the episode stayed politically contested, with U.S. Rep. Dina Titus and a retired firefighter publicly questioning the rescission[31][40].

A broader safety and environmental pattern

The firefighter case sits alongside others. A February 2024 Nevada OSHA action cited TBC for eight serious violations ($112,504) after workers reported wading through chemical sludge and ankle-deep water; the agency said 15–20 employees were burned connecting hoses without proper protective equipment[32]. In November 2025 Clark County moved to fine TBC nearly $500,000 for allegedly dumping drilling fluid into the county sewer system and "feigning compliance" with inspectors[33].

Key-person and political risk

TBC is hard to separate from Elon Musk. Its president, Steve Davis, served as Musk's de facto No. 2 at DOGE in 2025 before exiting with him — placing the company squarely inside the "Elon empire" and its political crosscurrents[34]. The Nashville council's 20–15 opposition vote and the politicization of the OSHA episode both show how local support for TBC now tracks sentiment toward Musk as much as the engineering[23]. Concentration cuts both ways: Musk's reach opens doors (Nashville, the LVCVA contract) and creates a single point of political failure.

Method

Research proceeded by fan-out web search and direct source fetching: company pages (boringcompany.com), Wikipedia entries for the company and the Vegas Loop, the Tennessee Governor's office releases, local press (Las Vegas Review-Journal, Nevada Current, Fox 17 Nashville, The Nevada Independent), investigative reporting (Fortune), transit-expert analysis (Jarrett Walker / Human Transit), and analyst/secondary write-ups (Sacra, Contrary Research, Capitaly). Every cited URL was fetched during the research run; each claim carries a source id, a tier, a confidence flag, and a stance.

Source mix

The Boring Company is US-based and English-speaking, so all sources are English-language; no native-language pass was required. The stance mix was kept deliberately balanced so that the supporting and critical evidence on each question are both represented.

Neutrality commitment

This study is a compilation that lets you reach your own conclusion, not an argument for or against The Boring Company. Each section presents the strongest version of both the "disruption" and the "vaporware" readings, attributes critical claims to named sources, and separates sourced fact from labeled analysis.

Cross-checked at build time by an automated link checker. A few sources may be bot-walled and return HTTP 403/429 to automated fetchers; they were read during the research run and are labeled accordingly. See Methodology & Limits.