Mutual assured destruction (MAD) is a doctrine of military strategy and national security policy which posits that a full-scale use of nuclear weapons by an attacker on a nuclear-armed defender with second-strike capabilities would result in the complete annihilation of both the attacker and the defender. It is based on the theory of rational deterrence, which holds that the threat of using strong weapons against the enemy prevents the enemy's use of those same weapons. The strategy is a form of Nash equilibrium in which, once armed, neither side has any incentive to initiate a conflict or to disarm.

The result may be a nuclear peace, in which the presence of nuclear weapons decreases the risk of crisis escalation, since parties will seek to avoid situations that could lead to the use of nuclear weapons. Proponents of nuclear peace theory therefore believe that controlled nuclear proliferation may be beneficial for global stability. Critics argue that nuclear proliferation increases the chance of nuclear war through either deliberate or inadvertent use of nuclear weapons, as well as the likelihood of nuclear material falling into the hands of violent non-state actors.

The term "mutual assured destruction", commonly abbreviated "MAD", was coined by Donald Brennan, a strategist working in Herman Kahn's Hudson Institute in 1962. Brennan conceived the acronym cynically, spelling out the English word "mad" to argue that holding weapons capable of destroying society was irrational.

Under MAD, each side has enough nuclear weaponry to destroy the other side. Either side, if attacked for any reason by the other, would retaliate with equal or greater force. The expected result is an immediate, irreversible escalation of hostilities resulting in both combatants' mutual, total, and assured destruction. The doctrine requires that neither side construct shelters on a massive scale. If one side constructed a similar system of shelters, it would violate the MAD doctrine and destabilize the situation, because it would have less to fear from a second strike. The same principle is invoked against missile defense.

The doctrine further assumes that neither side will dare to launch a first strike because the other side would launch on warning (also called fail-deadly) or with surviving forces (a second strike), resulting in unacceptable losses for both parties. The payoff of the MAD doctrine was and still is expected to be a tense but stable global peace. However, many have argued that mutually assured destruction is unable to deter conventional war that could later escalate. Emerging domains of cyber-espionage, proxy-state conflict, and high-speed missiles threaten to circumvent MAD as a deterrent strategy.

The primary application of this doctrine started during the Cold War (1940s to 1991), in which MAD was seen as helping to prevent any direct full-scale conflicts between the United States and the Soviet Union while they engaged in smaller proxy wars around the world. MAD was also responsible for the arms race, as both nations struggled to keep nuclear parity, or at least retain second-strike capability. Although the Cold War ended in the early 1990s, the MAD doctrine continues to be applied.

Proponents of MAD as part of the US and USSR strategic doctrine believed that nuclear war could best be prevented if neither side could expect to survive a full-scale nuclear exchange as a functioning state. Since the credibility of the threat is critical to such assurance, each side had to invest substantial capital in their nuclear arsenals even if they were not intended for use. In addition, neither side could be expected or allowed to adequately defend itself against the other's nuclear missiles. This led both to the hardening and diversification of nuclear delivery systems (such as nuclear missile silos, ballistic missile submarines, and nuclear bombers kept at fail-safe points) and to the Anti-Ballistic Missile Treaty.

This MAD scenario is often referred to as rational nuclear deterrence.