In Global Catastrophic Risks 2016, we referred to a number used in the Stern Review on the Economics of Climate Change: a 0.1% annual chance of human extinction. Stern uses this as a modelling assumption for discussing discount rates. There is a small amount of discussion of this figure in the Stern Review. It is clear that Stern did not intend the figure as an estimate. We’ve had a critique of our use of the figure forwarded to us, and we think its analysis is useful. We had no intention of using this figure in a misleading way, and we agree that we made a mistake in how we presented this figure. We should have been clearer about what the status of the number in the Stern Review was and about how we intended to use the comparison.
Throughout the rest of the report, we are very explicit that we do not believe it is possible to make robust probability estimates of extinction or catastrophic risk and do not attempt to (except for asteroid and super-volcano risk). This mistake does not affect the validity of the main points of the report – that global catastrophic risks are worth addressing and that there are things we can do to address them.
In our report, we originally wrote that:
“It is easy to be misled by the apparently low probabilities of catastrophic events. The UK’s Stern Review on the Economics of Climate Change suggested a 0.1% chance of human extinction each year, similar to some rough estimates of accidental nuclear warfare. At first glance, this may seem like an acceptable level of risk.
Moreover, small annual probabilities compound significantly over the long term. The annual chance of dying in a car accident in the United States is 1 in 9,395. However, this translates into an uncomfortably high lifetime risk of 1 in 120. Using the annual 0.1% figure from the Stern Review would imply a 9.5% chance of human extinction within the next hundred years.”
We were aware that the Stern Review used this figure merely as a modelling assumption, and were trying to give a concise accurate statement. Our intention in using the figure from the Stern Review was not to try to pin down an accurate estimate of the likelihood of global catastrophe, but to demonstrate that existing serious analysis treats the 0.1% probability as a plausible modeling assumption, which would have consequences that are interesting and non-intuitive.
We also had a full-page summary pull-quote, which said:
“The UK’s Stern Review on the Economics of Climate Change suggested a 0.1% chance of human extinction each year. If this estimate is correct, a typical person is more than five times as likely to die in an extinction event as in a car crash.”
This implies more confidence in the 0.1% figure than either we felt or expect the Stern Review to have felt, and more than our argument required.
The car crash comparison was picked up in The Atlantic, which reported it as an unconditional claim and emphasised it in their article. We did not intend to argue that the 0.1% figure was an accurate estimate of extinction risk (as we did not plan to offer an estimate of extinction risk), so this was inadvertently misleading to Atlantic readers. We believe that in general The Atlantic stood out by doing an excellent job of engaging constructively with our work.
We are also sorry in particular that we allowed the word ‘estimate’ to enter the soundbite on the full page. This error occurred at a late stage in the editing; the word was introduced to avoid an ambiguity, but not subjected to proper review.
We have carefully reviewed our language concerning the Stern Review, and written to our partners at the Global Challenges Foundation who published the report to change this to:
“The probabilities of these catastrophic events are low but not negligible. Moreover, small annual probabilities compound significantly over the long term.
We do not know of a robust estimate of the annual probability of global catastrophic risk. Nor do we believe that we are able to create a robust estimate because the uncertainties in key parameters are so large. However, for extinction risks some experts have suggested that a 0.1% annual chance of extinction is within the range of plausible orders of magnitude. A 2008 Oxford survey of expert judgement on the topic implied an average annual extinction risk over the next century of around 0.2%.  The UK’s Stern Review on the Economics of Climate Change used 0.1% as an upper bound modeling assumption for annual extinction risk.
Now let’s suppose that the chance of extinction were 0.1% per year and consider the consequences. It may seem at first glance that this would be an acceptable level of risk. However, that would mean an individual would be more than five times as likely to die in an extinction event than a car crash. Moreover, these small annual probabilities add up, so that the chance of extinction within the next century under this scenario is 9.5%. A global catastrophe, which involves the death of 10% of the global population, is more likely than an event that involves human extinction. As a result, even if 0.1% were on the high side for extinction risk, it might be of the appropriate order of magnitude for global catastrophic risk.”
We are also correcting a citation and adding a citation to
 Sandberg, A. & Bostrom, N. (2008): “Global Catastrophic Risks Survey”, Technical. Report #2008-1, Future of Humanity Institute, Oxford University: pp. 1-5.
We are also making the corresponding changes in the one-page soundbite and will also write to The Atlantic to inform them of the inadvertent inaccuracy in the article, and offer to help in correcting the nuance of the article.