How Does Filecoin Work?

Proof of Spacetime (PoST) — storage providers must periodically prove they are continuously storing client data over time by submitting cryptographic proofs (WindowPoST) at regular intervals, with failure to prove resulting in slashing of collateral

Genuinely novel cryptographic construction. Unlike proof-of-work or proof-of-stake, PoST proves useful work (data storage) over time. Complex SNARKs make verification efficient but the proof generation is computationally intensive and requires specialized hardware.

Proof of Replication (PoRep) — storage providers must prove they have created a unique, dedicated physical copy of the data via a slow encoding process (sealing), preventing Sybil attacks where one copy serves multiple deals

Another genuinely novel construction. The sealing process is intentionally slow and hardware-intensive to ensure providers cannot fake unique replicas. This creates high hardware barriers to entry (GPU sealing rigs) which contributes to centralization among well-capitalized providers.

Expected Consensus (EC) — block producers are elected probabilistically based on their proportion of quality-adjusted power (raw storage weighted by deal type), with multiple block winners possible per epoch (tipsets) to improve throughput

EC is a variant of proof-of-stake where the stake is storage power rather than tokens. The tipset mechanism (multiple blocks per epoch) is unusual but not fundamentally novel. The quality-adjusted power weighting incentivizes verified deals over committed capacity.

Storage deal market — clients and storage providers negotiate deals on-chain specifying data size, duration, price, and collateral; the built-in storage market actor mediates deal creation, payment escrow, and dispute resolution via the proof system

On-chain marketplace for storage deals. In practice, most storage deals are heavily subsidized by Filecoin Plus DataCap allocations (free storage for verified clients), meaning the market does not function as a true price discovery mechanism.

FIL collateral and slashing — storage providers must lock FIL as collateral proportional to their committed storage; failure to maintain sectors (missed WindowPoST), early sector termination, or detected faults result in progressive collateral slashing

Standard collateral-slashing mechanism but with uniquely high requirements. Initial pledge requirement includes a storage pledge (projected block rewards) and a consensus pledge (circulating supply fraction). The high collateral requirements create significant barriers to exit and lock up large amounts of FIL supply.

Storage proof verification cost — the computational overhead of generating and verifying WindowPoST proofs every 24 hours creates a persistent operational cost for providers; as FIL price declines, the cost of proof generation hardware can exceed block reward value, making mining uneconomical

Miner collateral lock-up creating exit barriers — storage providers must lock substantial FIL collateral for the full sector lifetime (up to 540 days); if FIL price drops or storage economics deteriorate, providers face a choice between eating losses or terminating sectors and accepting slashing penalties, creating a slow-motion exit trap

Deal market manipulation via DataCap — Filecoin Plus notaries allocate free DataCap to verified clients, which provides 10x quality-adjusted power multiplier; this creates incentives to game the notary system for power advantages rather than serve real storage demand, distorting the storage market

Sector sealing hardware requirements — the PoRep sealing process requires expensive GPU rigs and substantial memory, combined with high collateral requirements this creates enormous capital barriers to entry, concentrating storage power among well-funded operations primarily in China and North America