The short answer: not natively
Chapter 18 already laid out the mechanical difference between the two consensus models: proof-of-work earns the right to add the next block by spending real, external electricity and hardware; proof-of-stake earns it by locking up the network's own coin as collateral. Bitcoin has run on proof-of-work since block zero and has no protocol-level concept of "stake" at all — there's no built-in mechanism where locking BTC on Bitcoin's own chain earns you a share of new issuance or transaction fees the way locking ETH does on Ethereum. If a product, exchange, or article uses the word "staking" for Bitcoin specifically, it's describing something built on top of or alongside Bitcoin, never a change to Bitcoin's own base layer.
Then what is everyone actually doing?
Three distinct arrangements get called "Bitcoin staking" in practice, and they don't share much beyond the label:
How the native version actually works
Case 3 is worth walking through mechanically, because it's the newest and least understood of the three, and because it doesn't require trusting a custodian the way the other two do. A staker creates a Bitcoin transaction that locks their BTC into a special output with two conditions built into the script: a timelock (the staker can reclaim the funds themselves after a set period, using their own key — nobody else can touch it before then) and a slashing path, which only activates under a specific, provable form of misbehavior.
The staker delegates to a "finality provider" — an operator, conceptually similar to a validator, who actually participates in securing the other chain. If that operator ever double-signs (dishonestly signs two conflicting blocks at the same height, the same kind of provable misbehavior every proof-of-stake system needs to punish), the cryptographic signature scheme used mathematically reveals the operator's private key the moment they do it. Anyone can then use that exposed key to execute the slashing transaction — but critically, what typically gets slashed under a well-designed version of this mechanism is a pre-agreed, usually partial, fraction of the delegation, not automatically the staker's entire principal, and the exact terms vary by implementation and by which finality provider was chosen. The staker's coins never leave Bitcoin's own chain to be at risk of a bridge hack; the entire arrangement lives in a Bitcoin transaction's own script.
Why yields aren't a fixed number
Every version of "Bitcoin staking" advertises a yield, and none of those numbers are a fixed, guaranteed rate — they move with how much total BTC is competing for the same delegation opportunities, market demand for the underlying service, and (for the lending-relabeled version) whatever interest rate the platform can actually charge borrowers at any given moment. This book doesn't publish live rates for the same reason it doesn't publish a live BTC/USD price: any specific percentage printed here would be stale before it mattered. The category-level distinction — custodial loan, wrapped-and-bridged, or native self-custodied delegation — is the durable thing worth understanding; the exact APY on any of them today is not.
The custody question, again
Notice that two of the three cases above put your BTC back in someone else's hands — the same tradeoff Chapter 13 already covered for exchanges and Chapter 8 covered as the one rule that outranks every other rule. "Staking" doesn't exempt an arrangement from that question; it just adds a yield on top of the same custodial risk, which arguably makes it more important to ask who actually holds the keys while your coins are locked, not less. Only the native, self-custodied case avoids handing your keys to anyone — and even there, you're still trusting a finality provider's honesty and a young protocol's code, just not trusting them with your private key directly.
Bitcoin was never built to be staked. Everything called "Bitcoin staking" is something else entirely, wearing the one word that makes it sound familiar.