According to News Channel 5 Nashville, Tennesseans lost more than $15 million to Bitcoin ATM scams in 2025, and victims are still pursuing some form of recourse. The report says this fight is ongoing, though the feed summary gives no breakdown of how the losses were distributed across machines, operators or victim age groups. That is the whole of what we know from the summary; the rest is context worth adding.
Why this matters for people using cash machines
Bitcoin ATMs occupy a strange corner of the payments world: cash goes in, coins move out, and the transaction is final the moment it confirms on-chain. There is no bank to call, no chargeback form, no fraud department that can claw money back once a victim has been walked through a machine by a scammer on the phone. That structural fact is the actual story behind every headline number like Tennessee’s $15 million, and it is why these losses concentrate so heavily among people persuaded under pressure — a fake IRS agent, a romance-scam partner, a tech-support caller — rather than among people making informed investment choices. The machine itself is rarely the fraud; it is the fastest legal on-ramp a scammer can point a victim toward.
It also matters because the fee structure of this industry has never been transparent to the people most likely to be targeted. The Federal Reserve Bank of Kansas City’s 2023 research put the median self-reported buy fee at 16 percent and sell fee at 15 percent, with spread adding another 5 to 7 percent on top — and noted that operators charging above average tend to stop reporting, which biases even that number low. A retiree told to feed $10,000 into a machine to “protect” their accounts is not just being defrauded by a stranger on the phone; they are also absorbing a fee structure that would raise eyebrows in almost any other cash business, on top of the fraud itself.
Our view
We think state-level reporting of dollar totals is useful but is being treated as if it were the whole picture, and it isn’t. A single statewide figure tells you nothing about which operators’ machines are actually being used in these scams, whether certain locations or kiosk networks show disproportionate involvement, or whether the operators involved are cooperating with law enforcement or slow-walking subpoenas. We have seen, in building our own directory, how uneven this industry’s basic bookkeeping is — roughly one in five listings we checked during build-out no longer existed by the time we verified them. An industry that can’t keep its own location data current is not one that should be trusted to self-police fraud exposure, and state totals without operator-level detail let every operator hide behind the average.
We also think it’s fair to say Tennessee authorities and reporters are late to a problem that has been visible in fee and complaint data for years. The Kansas City Fed’s fee research is now over two years old and flagged the reporting bias explicitly; state regulators had the receipts to act on licensing and consumer-warning requirements well before 2025’s losses accumulated. Chasing victims’ fighting-back stories after the fact is worthwhile journalism, but it is not a substitute for the harder regulatory work of forcing operator-level transparency before the losses happen.
What to watch
Watch whether Tennessee, or any state citing large loss totals, follows up with operator-specific data — which company’s machines were used, how many transactions, what the licensing status of those machines was at the time. Also watch what happens to fraud-loss reporting following industry contraction: Bitcoin Depot’s Chapter 11 filing in May 2026 took more than 9,000 machines offline nationally, and a shrinking, consolidating fleet changes both the opportunity for fraud and the incentive for remaining operators to cooperate with state investigators rather than compete on lax compliance. If loss totals keep climbing even as machine counts fall, that is the signal that this is a targeting problem, not a machine-count problem.
