Briefing
The core research problem is the critical degradation of blockchain security due to selfish mining, a threat amplified by adverse network conditions such as rushing, varying block rewards from Maximal Extractable Value, and petty-compliant miners. The foundational breakthrough is the Mutual-Assured-Destruction Directed-Acyclic-Graph, or MAD-DAG, which introduces a novel ledger function. This function dictates that if a chain-length tie occurs, the content of all tied blocks is discarded, preventing the block proposer from collecting transaction fees or MEV. This mechanism front-loads a cost on deviation, forcing a selfish miner to achieve a strictly longer chain to profit, thereby negating the economic advantage gained from rushing and MEV, and ensuring protocol security under practical latency.
Context
Before this research, established Proof-of-Work protocols were theoretically secure only under idealized assumptions, specifically constant block rewards and limited network rushing. Foundational selfish mining models demonstrated that a rational adversary required a minimum of 25% of the network’s hash power to profit. However, the introduction of high-value MEV and the reality of fast block propagation (rushing) shattered this theoretical threshold. This created an unsolved foundational problem where the security bound degraded to as low as zero percent, meaning rational deviation was always profitable and leading to a fundamentally unstable consensus environment.
Analysis
MAD-DAG’s core mechanism is the “MAD ledger function,” a radical departure from traditional longest-chain rules that simply select one of the tied chains. The logic is simple → mutual destruction of value upon conflict. A selfish miner’s primary strategy is to release a private chain when its length equals the public chain’s, hoping for a tie-break that grants them the block reward and MEV.
The MAD ledger function removes the incentive by making the tie-break yield a contentless block, meaning zero transaction fees and zero MEV. This forces the attacker to expend significantly more resources to mine a strictly longer chain, raising the profitability threshold back to a secure level even when block rewards vary due to MEV.
Parameters
- Security Threshold Degradation → 0% → The minimum mining power required for profitable selfish mining under adverse conditions (before MAD-DAG).
- MAD Ledger Function Action → Discard Block Content → The consequence of a chain-length tie in the MAD-DAG protocol.
- Mitigated Conditions → Rushing, MEV, Petty-Compliant Miners → The three adverse, real-world conditions MAD-DAG is designed to secure consensus against.
Outlook
This research opens a new avenue for designing robust consensus protocols that internalize and neutralize the economic incentives of MEV at the foundational layer, rather than relying on application-layer mitigations. The strategic implication is a shift toward “destruction-based” mechanism design, where the profitability of adversarial strategies is eliminated by destroying the extracted value upon detection of a fork. In the next 3-5 years, this principle could be adapted to Proof-of-Stake protocols to secure transaction ordering and validator incentives, potentially leading to provably MEV-resilient base layers.
Verdict
The MAD-DAG protocol provides a critical, theoretically sound mechanism design solution that restores the economic security of blockchain consensus under the existential threat of Maximal Extractable Value.
