Can the same compute securing blockchains also run AI workloads?
Miners get paid for securing networks. AI providers get paid for inference. Proof-of-Useful-Work tries to merge both.
Early PoUW designs failed because useful work was hard to verify, adjust difficulty for, and couple tightly to consensus. Recent progress makes it more credible.
Bullish case: AI demand funds blockchain security.
AI customers pay for compute. Token rewards subsidize costs, lowering prices, attracting more demand and supply. Classic flywheel.
But here's the catch: AI demand doesn't automatically secure the chain. It only works if useful work is part of consensus and revenue stays tied to the canonical chain. Otherwise, miners just take their GPUs elsewhere.
More customer revenue means less reliance on token emissions. The AI business can scale without token demand growing. Emissions might bootstrap a compute market that can't survive on real demand alone.
PoUW only becomes a flywheel if AI demand actually strengthens the blockchain AND can replace token subsidies without bleeding compute away from security.
Most PoUW projects will fail this test.
Miners get paid for securing networks. AI providers get paid for inference. Proof-of-Useful-Work tries to merge both.
Early PoUW designs failed because useful work was hard to verify, adjust difficulty for, and couple tightly to consensus. Recent progress makes it more credible.
Bullish case: AI demand funds blockchain security.
AI customers pay for compute. Token rewards subsidize costs, lowering prices, attracting more demand and supply. Classic flywheel.
But here's the catch: AI demand doesn't automatically secure the chain. It only works if useful work is part of consensus and revenue stays tied to the canonical chain. Otherwise, miners just take their GPUs elsewhere.
More customer revenue means less reliance on token emissions. The AI business can scale without token demand growing. Emissions might bootstrap a compute market that can't survive on real demand alone.
PoUW only becomes a flywheel if AI demand actually strengthens the blockchain AND can replace token subsidies without bleeding compute away from security.
Most PoUW projects will fail this test.
