The Great Filter is a proposed reframing of the Fermi Paradox, introduced by Robin Hanson in his 1998 essay The Great Filter - Are We Almost Past It?.
The development of space-framing intelligent life requires many steps to occur in sequence, such as the emergence of single-celled life and the transition from unicellular to multicellular life forms. Since we have not observed intelligent life beyond our planet, Hanson argues that there seems to be a developmental step that is so difficult and unlikely that it "filters out" nearly all civilizations before they can reach a space-faring stage — a "great filter".
From Hanson's essay:
Humanity seems to have a bright future, i.e., a non-trivial chance of expanding to fill the universe with lasting life. But the fact that space near us seems dead now tells us that any given piece of dead matter faces an astronomically low chance of begating such a future. There thus exists a great filter between death and expanding lasting life, and humanity faces the ominous question: how far along this filter are we?
If there is a "Great Filter", then this filter might be a step in our evolutionary past, in which case our civilization has already passed it.
But the hard step might also be ahead of us: surviving the creation of AGI, future biotechnology, nanotechnology, or some unknown risk. In that case, we should be worried, as the Great Filter seems to have been successful in stopping the development of every other civilization so far.
This suggests that estimating the location of the Great Filter may be important for helping estimate the magnitude of existential risk. Many efforts have been made in that direction, but much remains uncertain.
If we discovered traces of life on other planets, this would count as evidence for a later Great Filter. Finding that complex life had evolved independently both on Earth and some other nearby planet, would suggest that getting to such a developmental stage was relatively easy. Thus the Great Filter would almost certainly have to be at a later stage.
The study of past mass extinctions and astrobiology can provide ideas for estimating the location of a Great Filter. However, there are many difficulties involved. For instance, the time that it takes to pass a step doesn't reveal much about how easy or hard that step was. Robin Hanson gives the following example in his paper:
… [S]ay you have one hour to pick five locks by trial and error, locks with 1,2,3,4, and 5 dials of ten numbers, so that the expected time to pick each lock is .01,.1, 1, 10, and 100 hours respectively. Then just looking at those rare cases when you do pick all five locks in the hour, the average time to pick the first two locks would be .0096 and .075 hours respectively, close to the usual expected times of .01 and .1 hours. The average time to pick the third lock, however, would be .20 hours, and the average time for the other two locks, and the average time left over at the end, would be .24 hours. That is, conditional on success, all the hard steps, no matter how hard, take about the same time, while easy steps take about their usual time.
In a subsequent paper, Hanson constructs a simulation of the distribution of the hard steps, which suggests that there should be about four to seven hard steps, uniformly distributed in our past — a series of lesser filters, rather than a single "great" one.
Hanson's follow-up also suggests that there has been at least one hard step since the evolution of hominids, and that the best extinction model that fits all these requirements is William Schopf's model.
Taking evolutionary arguments for AGI and observation selection effects together, Bostrom and Shulman argue that Hanson’s results can help estimate the difficulty of creating AGI.