Cause X is a cause area currently neglected by the effective altruism community, typically due to some form of moral blindness or fundamental oversight, yet more important than all the causes currently prioritized by it.
The idea was introduced by William MacAskill,[1] possibly by analogy with Derek Parfit's "Theory X", a currently unknown hypothetical theory that would solve some important open problems in population ethics.[2] ("Cause X" is also sometimes used loosely to refer to any promising and neglected cause,[3][4] but this is not how the expression is generally understood.)[5]
The existence of Cause X should not be taken for granted; whether there is a Cause X is an open question. In contrast to Parfit's Theory X, however, there are no impossibility theorems that could disprove the existence of Cause X.[6] Furthermore, if there is a Cause X, there is no determinate way of ascertaining that it has been found; this is another disanalogy with Theory X, whose conditions are defined with sufficient precision that it is possible to establish when a theory satisfies them.
Still, there appear to be good reasons for expecting Cause X to exist. All previous generations overlooked highly important causes—usually by neglecting large groups of morally relevant beings, such as women, racial minorities, and nonhuman animals—,[7] so it would be a remarkable coincidence if our generation were the first to avoid this moral shortcoming.[1][8][9] In addition, the appearance that the present generation is different may itself be debunked as a manifestation of the "end of history illusion",[10] "new era thinking",[11] and related cognitive biases. Furthermore, the space of possible cause areas is vast, and humans have only recently begun to explore it systematically, so it seems antecedently very likely that a cause more important than the current top causes remains to be found.[8] Yet another argument for the existence of Cause X is that the effective altruism community has, to a certain extent, changed its views about which causes are most important over the years. An induction from this history suggests further changes in what causes the community will consider most impactful.
Some heuristics for finding Cause X have been proposed.[3][12] Kerry Vaughan suggests three such heuristics:
Heuristic 1: Moral circle expansion. One apparently robust historical trend since at least the last few centuries is the gradual expansion of the circle of moral concern.[13][14][15] Furthermore, this expansion appears to account for much of the moral progress during this period. Thus, a plausible heuristic is to push this expansion even further. This heuristic suggests wild animal welfare, invertebrate welfare, artificial sentience, as well as research on moral patienthood, as Cause X candidates.
Heuristic 2: Transformative technology. Technological progress has the potential to radically transform the world in morally relevant respects. This assessment is plausible both based on historical analysis—many of the most significant changes have occurred due to some form of human innovation—and upon consideration of various anticipated technologies not yet developed, such as whole brain emulation, artificial general intelligence, and atomically precise manufacturing. The heuristic suggests working on making these technologies safer, such as AI safety, as well as differential technological development, as Cause X candidates.
Heuristic 3: Crucial considerations. A crucial consideration is one that warrants a major reassessment of a cause's impact. Actively looking for such considerations is thus of clear relevance for finding Cause X. Here, the heuristic would favor making lists of crucial considerations, as well as the search for additional "deliberation ladders" in existing arguments for specific causes.
The meta-heuristic of holding events where attempts are made to find potential Cause X candidates has also been suggested as an effective discovery method.[16] Such events could take the form of informal meetups, conference workshops, or academic conferences.