First, yes, the unit of selection is of course the individual.
I'd argue it's the gene.
But a species-wide feature that causes each individual's chance of survival and propagation to go down does, in fact, imperil the survival of the species when it is selected against in every individual's case. When every individual dies, so does the species; similarly, if merely enough individuals die that there's a genetic bottleneck, the species also risks extinction because of inbreeding-related unfitness of individuals.
To the extent that this works against runaway selection pressure for intelligence, it also works against runaway selection pressure in general, which as you say, still happens a lot.
Next time try actually thinking through what is being said, instead of partial-pattern-matching in an effort to score points.
Please tell me why you think I was doing this, if you've good reason for not assuming good faith then I'll try to avoid giving this impression in future.
And yes, runaway selection pressure in general happens a lot. Now, how often does it happen specifically on the characteristic of intelligence, instead of all the other possible features it could happen on? Looking at the only paleontological record we have available, it seems to be awfully rare.
Fair point. I'm not sure if they'd count as runaway selection pressure, but aren't there lots of examples for animal-level equivalents of social intelligence? If so, then that counts as it looks like that's the main reason we got smart.
I'd argue it's the gene.
Fair enough. The point is, nothing I said depended on the premise that the species as the unit of selection, it was merely stating the effect that selection (on whatever unit you like) had on the species. These are distinct concepts. Nothing I said was dependent on the erroneous idea of the species as the unit of selection.
The most charitable interpretation of your "cough" link was that you had read my statement, saw I wasn't making the mistake addressed in your link, but thought I should have made a big verbose sho...
One serious issue for evaluating existential risk is working out whether most of the Great Filter is behind us or in front of us. This relates to the Drake Equation and similar attempts to estimate the frequency of life in an obvious way.
Over the last few years, it has become increasingly apparent that extrasolar planets are common. However, what fraction of these planets lie in their stars habitable zone has still been an open question, primarily because most of our current methods for planet finding easily find planets that are either very large or are very close to their star (ideally both).
A new study, using the data from the Kepler spacecraft, estimates that about a third of all stars similar to the sun have at least one planet in the habitable zone. There are some issues with this estimate, and Phil Plait discusses them at his blog. The estimate has a large amount of variance. The paper actually estimates 34% +/- 14% and the issues that Phil brings up increases the uncertainty in both directions but it seems safe at this point to consider this not being very far off.
One obvious issue from a Fermi perspective is that some systems will likely have multiple planets in this zone. Also, having planets in the habitable zone is clearly not sufficient for life. By the standard estimates for habitable zones, Venus and Mars are both in the habitable zone of the sun. And there may very well be ways for life to arise outside the habitable zone. Moons like Europa and Titan seem to be excellent candidates, and we can't rule out more exotic forms of life in other habitats although that seems not too likely right now.
However, one thing this makes clear: The part of the Great Filter that is behind us that is due to planets not lying inside the habitable zone is small. So the question is, what does this mean for our estimates of how much of the Filter is behind us and how much is in front of us?