Fair equality of opportunity is stronger than formal equality of opportunity. Formal equality requires only that positions be legally open to all — that no one be excluded from competition by law or official policy. Fair equality requires something more: that the background institutions of society — the educational system, the family structure, the economic arrangements — do not produce differences in life prospects between individuals of similar talent and motivation based merely on the accident of their social starting position. Fair equality is demanding because it looks beyond the formal rules of competition to the conditions under which competition actually occurs, and it holds those conditions to a standard of justice. Applied to the AI transition, fair equality of opportunity requires that access to AI tools and the benefits flowing from them not be determined by the accident of where one happens to be born, what family one happens to be born into, or what educational infrastructure one happens to have access to.
The principle is given strict lexical priority over the difference principle within Rawls's second principle. An arrangement cannot satisfy the difference principle if it violates fair equality of opportunity, regardless of how much it benefits the least advantaged. This ordering has sharp consequences for the AI transition. Even if a given distribution of AI's gains benefits those at the bottom more than any alternative distribution would, the arrangement is unjust if the access to the tools and the capacity to convert that access into substantive capability are systematically determined by social starting position.
The AI transition has simultaneously narrowed the gap between formal and substantive equality in certain dimensions and widened it in others. The narrowing is real: AI tools have lowered the floor of who gets to build. A person with an idea and a subscription can now produce working software that would have required a team twenty-four months ago. But the narrowing occurs along one dimension — access to productive tools — while leaving other dimensions untouched or exacerbated. Formal equality of access — same tool, same subscription price — coexists with profound substantive inequality in the conditions under which the tool is used.
Amartya Sen's distinction between formal and substantive capability sharpens this point. A person who formally has access to an AI tool but lacks the infrastructure, the training, the stability, and the social support to use it effectively does not possess the capability that the access is supposed to provide. The capability is nominal. And nominal capability, evaluated by the standard of justice, does not satisfy fair equality of opportunity. The parallel to the public library system is precise: building the library is necessary but not sufficient. Fair equality of opportunity is not achieved by making AI tools available; it is achieved by building the educational, infrastructural, and social conditions that allow all citizens to benefit from the tools.
The institutional requirements are substantial. Educational reform must be designed from the perspective of the least advantaged students, not from the perspective of the educational institutions. Infrastructure investment must be directed toward the communities with the least access. Language model development must account for the linguistic and professional diversity of the global population. Thomas Pogge's extension of Rawls to the global level argued that the institutional arrangements governing international economic relations impose duties on wealthy nations to reform structures that produce and perpetuate global poverty — a framework directly applicable to the AI transition, where the nations and corporations that capture the largest share of the gains bear corresponding duties to invest in the conditions that would allow the globally disadvantaged to benefit from the same tools.
Rawls developed fair equality of opportunity in A Theory of Justice (§§12–14) as the stronger conception he argued reasonable parties in the original position would choose over mere formal equality. The principle drew on the American tradition of "equal opportunity" while sharpening it philosophically — distinguishing the demanding substantive conception from the thin formal conception that had dominated American policy discourse.
Nussbaum's and Sen's capabilities approach extended Rawls's principle by specifying more concretely what conversion from formal access to substantive capability requires, and by identifying the specific capabilities that every person needs to lead a life of dignity. The capabilities approach is not simply a variation of Rawls's framework but a distinct position; its relationship to fair equality of opportunity remains a subject of active philosophical debate.
Stronger than formal equality. Formal equality requires legal non-exclusion; fair equality requires that background institutions not produce differences in prospects based on social starting position.
Priority over the difference principle. No arrangement can satisfy the difference principle if it violates fair equality of opportunity; the lexical ordering is strict.
Background justice. The principle evaluates the conditions under which competition occurs, not merely the rules of the competition itself.
Conversion capacity. Formal access to a resource does not constitute opportunity if the person lacks the conditions to convert access into substantive capability.
Global reach. In a world where institutional arrangements cross national borders, fair equality of opportunity cannot be satisfied by purely domestic measures; the AI transition is inherently global in scope.
The principle's demandingness has been contested from multiple directions. Libertarians argue that it permits state intervention in private family life and educational choice that violates individual liberty. Communitarians argue that it presupposes an atomized conception of social membership incompatible with the particular attachments through which people actually come to have talents and motivations. Cohen argued in the other direction that even fair equality of opportunity is insufficient — that true equality requires attention to inequalities in natural talents themselves, which Rawls treated as morally arbitrary but permitted to generate differential rewards.