The Energy Is the Point

Mining has moved through three eras, and may be entering a fourth:

• Individuals (2009–2013). CPUs, then graphics cards, run by hobbyists in bedrooms. Fully decentralized — and tiny.
• Industrial (2013–now). Purpose-built ASIC chips and economies of scale concentrated mining into big firms and a handful of pools. Powerful — and a real centralization concern for a network whose whole value is decentralization.
• The AI pivot (now). With margins squeezed after the 2024 halving, some of the biggest miners are redirecting capacity toward AI compute, which currently pays better. Rigs, power contracts, and datacenters are being repurposed.
• A possible re-decentralization (emerging). Cheap secondhand ASICs, home units that recapture heat, and new pool protocols that hand block-building control back to individual miners. If big players chase AI, they may leave room for many smaller hands to regain share.

Here is the single most counterintuitive fact in Bitcoin, and the honest version of this piece lives or dies on stating it plainly: lower energy use would make the network weaker, not healthier.

• Hashrate IS the security. The energy poured into mining is the wall protecting the ledger. To rewrite history or double-spend, an attacker must out-compute the entire honest network at once — the “51% attack.” The more total energy securing it, the more physically and financially impossible that becomes.
• So “use less energy” = “be easier to attack.” This is the one place where the green intuition runs exactly backwards. A Bitcoin secured by little energy is a Bitcoin that’s cheap to forge. The energy is not waste; it’s the cost of making the money unforgeable.

The critic’s number is real. The honest rebuttal has three layers, and none of them is hand-waving:

Layer 1 — it’s self-regulating, not a runaway

Mining energy isn’t fixed — it’s an equilibrium set by profitability. Miners add hashpower until the marginal cost of power roughly equals the marginal reward. Bitcoin doesn’t “want” infinite energy; it consumes exactly as much as its price justifies, and no more. If price fell, hashrate and energy would fall with it. The reason usage is high is that a high price makes mining profitable — the consumption is a symptom of value, not a runaway appetite.

Layer 2 — compared to WHAT?

“Uses as much as country X” is a scary absolute number that ignores what it buys and what it replaces. The honest comparison isn’t ‘Bitcoin vs. nothing.’ It’s Bitcoin vs. the entire physical apparatus of legacy finance — bank branches, card networks, ATMs, cash printing and armored transport, gold mining and vaulting. Bitcoin does the job of a global monetary system with a single shared ledger.

Layer 3 — the part that flips it: waste energy

Because miners chase the cheapest power, can operate anywhere, and shut off instantly, they increasingly consume energy that would otherwise be wasted:

• Landfill methane. Landfills leak methane — far worse than CO₂ as a greenhouse gas. Most is flared or vented because there’s no economic reason to capture it. A miner on-site turns that gas into revenue, which funds the capture that wouldn’t otherwise happen. You’re monetizing the destruction of a worse emission.
• Stranded & flared gas. Gas at remote or abandoned wells with no pipeline gets flared — burned for nothing — or vented. A miner is a buyer that shows up anywhere and turns wasted energy into money.
• Curtailed renewables. Wind and solar often produce more than the grid can take, and the excess is dumped. A miner soaking up that curtailed output makes the wind farm more financeable — mining can actually improve the economics of building clean energy.

The strongest single example lives in Texas. Because a miner can power down in seconds with no product ruined and no line to restart, it’s a perfectly interruptible load — and that makes it a grid asset, not just a grid burden.

• Miners shut off when people need the power. During heat waves and winter storms, Texas miners curtail — in seconds — and are paid to do it, freeing electricity for homes and hospitals exactly when the grid is stressed.
• They’re a buyer of last resort the rest of the time. By purchasing cheap off-peak and excess power, they give generators a reason to build capacity that’s there when demand spikes — then step aside when it does.

Here is what most Bitcoin energy pieces bury, and we won’t: Bitcoin mining and AI now compete for the same physical substrate — cheap power, datacenters, cooling, grid interconnects. That competition is real and it cuts both ways.

• When Bitcoin is profitable, miners bid up power and datacenter capacity — raising the cost floor for everyone, including AI.
• When AI is more profitable (arguably now), capacity flows to AI — and if Bitcoin recovers and pulls it back, AI compute gets more expensive, because both draw from one finite pool of cheap-power-plus-datacenter.
• The double-edge for security: the AI option helps miners survive the bear without dumping all their Bitcoin or going dark — which keeps hashrate up through the trough. But if AI pulls too much hashpower away for too long, that’s a genuine security concern. The healthy version is AI as a bridge through the bear, not a permanent exodus.

The dishonest energy defense says ‘Bitcoin uses only renewables and it’s all free.’ That’s not true and you don’t need it. The honest case is stronger:

• Concede: the network genuinely needs the energy — that’s the security. Hashrate can’t just fall without making Bitcoin easier to attack.
• Concede: mining has centralized, and the AI competition for compute is real and unresolved.
• And still: the energy is self-regulating, increasingly sourced from waste nobody else can use, and able to stabilize grids rather than strain them — while new tools push control back toward many hands. The honest version wins without pretending the hard parts away.