Invest1 publisher3 min readPublished
Aurora's 200-truck target tops out near $81 million of annual revenue
Aurora runs 20 driverless trucks today and wants 200 by December, but covering last year's $800 million loss with revenue alone would take closer to 2,000 of them at full utilisation.
The Investor · Invest desk
What happened
- Aurora Innovation has 20 driverless trucks on the road and plans to grow the autonomous fleet to 200 by the end of this year, running on I-45 and I-20 in Texas.
- Aurora lost more than $800 million in 2025 and is targeting positive free cash flow by 2028 on the back of a substantial increase in trucks on the road.
- Aurora has logged more than 440,000 miles on public roads since it began testing in 2022, most of them with a human safety operator behind the wheel.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
- cost Shippers keep the 45 cents a mile only while Aurora prices at 85 cents; if a mile costs more than that to supply, Aurora's shareholders are funding the discount.
- constraint The 2028 free-cash-flow date needs a fleet nearer 2,000 trucks than 200, so hitting December's target settles little about the date itself.
- decision Freight buyers now have to price the non-driving work uShip's Sean Wu describes, and decide whether a per-mile service that omits it is cheaper in practice.
- contradiction Morgan Stanley's 7.5x and Bank of America's 45 cents only sit together if human fleets are earning around 7 cents of profit a mile, which makes the multiple a statement about how thin trucking margins are.
Multiply the numbers in Aurora's own pitch. A driverless truck can roll up to 20 hours a day; federal rules cap a human driver's hours [6]. The speed Chris Urmson keeps being asked to defend is 65 or 70 mph [8]. Call it 1,300 miles a day for one truck, 260,000 a day across 200 [1]. Bank of America puts Aurora's service at about 85 cents a mile [5], so the full fleet at that ceiling bills roughly $221,000 a day, or about $81 million a year [2]. Aurora lost more than $800 million in 2025 [11].
Both inputs are upper bounds, so treat $81 million as the best case. Getting to $800 million of revenue at 85 cents takes about 941 million miles a year, and at 474,500 miles per truck-year that needs close to 2,000 trucks [6], ten times the December target [7]. That is revenue, before the cost of supplying it. CNBC did not report what a mile costs Aurora to run.
Morgan Stanley transportation analyst Ravi Shanker estimates that "an autonomous fleet should be nearly 7.5X as profitable as a human-driven fleet today" [4]. Bank of America's comparison is narrower, and it is a labour line: about 85 cents a mile for Aurora services against roughly $1.30 for human driver wages and benefits, before indirect labour costs [5]. The gap is 45 cents [4]. For that to multiply fleet profit by 7.5, with every cent of the saving dropping through, the human fleet has to be earning about 6.9 cents a mile today [5].
Price is not the only thing a shipper weighs. Sean Wu, CEO of the freight firm uShip, said he understands the appeal of a driverless truck for lowering costs but questions how many shipping firms will want to ditch their drivers [17]. "The drivers do more than just move the rig. There is judgement. There is customer service. There is protection," Wu told CNBC [12].
The installed base is the other denominator. Two hundred trucks against the roughly 3.5 million Class 8 semis the Department of Transportation counts is about one in 17,500 [13][9]. Aurora has logged more than 440,000 miles on public roads since 2022, most of them with a safety operator in the cab [10]. At the ceiling rate, a 200-truck fleet covers that entire history in under two days [8]. Urmson said "We put this through 15 million tests before we put it out on the road" [9].
Two things would make this reading too harsh. If Aurora's cost to supply a mile sits well below 85 cents, the 2028 free-cash-flow date arrives on a fleet far smaller than 2,000 trucks. And the $1.30 driver line moves: the driver shortage that has freight firms shopping for cheaper reliable service [15] is the same shortage that lifts wages, widening the gap without Aurora building anything. Kodiak AI, Gatik and Tesla are also developing autonomous vehicles [14]. Urmson said "We've been able to put out timelines and we've basically hit those timelines" [3].
What to watch
- Whether 200 driverless trucks are actually running by 31 December, and how many of them carry no safety operator.
- Whether Aurora's per-mile price holds at the 85 cents Bank of America assumes as volume grows.
- Daily miles per truck in commercial service against the 20-hour, 1,300-mile ceiling.