ignoranceprior

Karma: 273
• Alternatively, if the Covid-19 deaths in NY state go above 3,333 in the first week of April, that seems like it would also falsify the hypothesis. (NY state has fewer than one third the population of the UK.) Unfortunately I think this is >80% to happen.

On April 4, the death toll in NY state surpassed 3,333. As of April 10, there are 7,844 deaths.

• The “Rationalist prepper thread” was actually posted on January 28, not January 20.

• If the IFR is indeed .003% (the upper end of your range), then assuming the worst case scenario that 100% of the population of the UK gets infected eventually, only .003%*66.4 million = approx 2000 people will die total.

Would you consider the theory falsified if the death toll in the UK surpasses 2000?

• I’m confused why you assume that 36-68% of the population in the UK is infected. I thought, based on comments here, that those numbers were the output of a model that made highly optimistic assumptions about IFR, not an attempt at estimating the actual proportion of infections.

Do you think this is a realistic range for the proportion already infected in the UK?

• What is your personal point estimate or credible interval for IFR?

• Epidemiologist Neil Ferguson, who created the highly-cited Imperial College London coronavirus model, which has been cited by organizations like The New York Times and has been instrumental in governmental policy decision-making, offered a massive revision to his model on Wednesday.

Ferguson’s model projected 2.2 million dead people in the United States and 500,000 in the U.K. from COVID-19 if no action were taken to slow the virus and blunt its curve.
However, after just one day of ordered lockdowns in the U.K., Ferguson has changed his tune, revealing that far more people likely have the virus than his team figured. Now, the epidemiologist predicts, hospitals will be just fine taking on COVID-19 patients and estimates 20,000 or far fewer people will die from the virus itself or from its agitation of other ailments.

Ferguson thus dropped his prediction from 500,000 dead to 20,000.

Author and former New York Times reporter Alex Berenson broke down the bombshell report via Twitter on Thursday morning (view Twitter thread below).

“This is a remarkable turn from Neil Ferguson, who led the [Imperial College] authors who warned of 500,000 UK deaths — and who has now himself tested positive for #COVID,” started Berenson.

“He now says both that the U.K. should have enough ICU beds and that the coronavirus will probably kill under 20,000 people in the U.K. — more than 12 of whom would have died by the end of the year in any case [because] they were so old and sick,” he wrote.

Thoughts on this?

1. Is it better to buy hydroxychloroquine or regular chloroquine? The studies I’ve found suggest hydroxychloroquine is safer and more potent, but it is a bit more expensive.

2. How many days worth of the drug is it reasonable to buy per person?

3. How much should someone take per day and how should the dosage be timed?

4. Can someone confirm that the products you can find on reliablerxpharmacy.com when searching for “Lariago” (500 mg chloroquine as phos) and “OXCQ” (200 mg Hydroxychloroquine Sulfate) are the right things to buy? If not, is there any other reputable or semi-reputable source that sells the right product?

• Maybe birth rates will increase if there are massive quarantines, for the same reason birth rates are said to increase during natural disasters (???). Very uncertain. Just throwing this idea out there, since I’ve seen little discussion of it.

• Is the 5-10% global mortality prediction conditional on COVID-19 infecting >10% of the world, or unconditional?

What do you think of the prospects for antivirals like remdesivir to be tested and mass-produced? How much could they lower CFR?

Why do you think other predictions, such as those given by Metaculus 1, 2, 3 are much less pessimistic?

Do you think shorting the market is a good idea still?

• It confuses me that I seem to be the first person to talk much about this on either LW or EA Forum, given that there must be people who have been exposed to the current political environment earlier or to a greater extent than me.

This isn’t an answer to your historical question, but I would like to point out that an EA recently wrote up his thoughts on speech policing here on the EA Forum, and I recall some previous relevant discussions as well (example).

• Do any AI safety researchers have little things they would like to get done, but don’t have the time for?

I’m willing to help out for no pay.

I have a backgound in computer science and mathematics, and I have basic familiarity with AI alignment concepts. I can write code to help with ML experiments, and can help you summarize research or do literature reviews.

• I thought you were a negative utilitarian, in which case disaster recovery seems plausibly net-negative. Am I wrong about your values?