I apologize for my bad English, this is not my native language. And probably I will make some mistakes when posting.

For over 2 years I have been reading materials on the topic of AI Safety. I don't have the appropriate education, cognitive abilities, knowledge. I do not even have time to learn the language. So I didn't hope to do something useful myself.

But once I tried to systematize quotations from one show in order to understand when the experts represented there are waiting for AGI and how likely they consider the extinction of humanity.

I thought it would be interesting to do so with the rest of the experts.
In addition, I have already seen and studied with interest such collections of quotes. It seemed to me that the best thing I could do was try to do something similar.

Therefore, I began to collect quotes from people who can be attributed to the experts. It turned out to be harder than I thought.

I have compiled a table with quotes from more than 800 experts. I tried not to distort the opinion of forecasters and simply copied from sources, sometimes deleting or slightly editing. My edits can be recognized by square brackets :)

1) The first column of the table is the name of the expert.
2) The second column is the year of the forecast. The table is built in chronological order.
3) The third column is the predicted time for AGI. Unfortunately, most people did not speak directly about time and probability. Because of this, many quotes came out rather vague. For example, “Machines are very far from being intelligent” or “And we can reach it in a close time”.
4) The fourth column is an opinion about Takeoff Speed. About how much progress will be accelerated after create of AGI.
5) The fifth column is the expert's opinion about the future of mankind with AGI. Choosing a quote here was the hardest. Most of all I was interested in the risk of extinction or serious shocks due to AI, and I tried to provide quotes that most fully reveal this particular topic.
6) The sixth column indicates the source of the quote.

That is, to the right of the forecaster's name, you can find out the date of the given quotes, his opinion about the time of the creation of AI, about the intellectual explosion and about the future of humanity, as well as get acquainted with the source.

Of course, cases where the expert spoke on the topic of time, the speed of self-improvement and the influence of AI in the framework of one material are quite rare. Therefore many cells are left empty.

I had to give several quotes per person, sometimes they were separated for years and even decades.

Since all the quotes are given in chronological order, the opinions of some people are “scattered” in the table.

For example, Gwern spoke about the future of mankind in 2010, about the growth of AI in 2014 and about the forecasts for the emergence of AI in 2018.
However, you can simply use search.

In addition, sometimes one person has already made a certain forecast but later changed or expanded his opinion. I tried to take into account such quotes.

I also reviewed anonymous expert interviews and indicated them. If the general list of respondents was known, I cited them as well.

It was difficult to decide who should be considered an expert and what quotes should be included in the work.

I had to make controversial decisions. The table includes a lot of people who are entrepreneurs but may have insights on advanced research. There are several futurists and philosophers in the table. There are writers like Clark and Vinge, whose opinion seems important to me.

I have a version of this work without chronological separation, where the quotes are more grouped by name. Perhaps someone will be more convenient.

It is difficult to draw conclusions from the work. The absolute majority of experts did not talk about exact dates and did not indicate the probability of their predictions.

I can only say that most forecasters do not expect AI in the near future, do not expect IE and seem optimistic.

In addition, it seemed to me that in the twentieth century the leading experts were on average more pessimistic: Turing, Wiener, I. J. Good, Fredkin, Shannon, Moravec, etc.

Young researchers are on average more optimistic than older ones - even in the field of AI Safety, where on average there are naturally more concerned people.

I think that to confirm almost any views you can find the opinion of a respected expert.

I really hope that for someone my work will be useful and interesting. Criticism and additions are welcome.

New to LessWrong?

New Comment
2 comments, sorted by Click to highlight new comments since: Today at 6:57 PM

This spreadsheet is super impressive and has been very useful to me (it allowed me to find some very interesting stuff, like this discussion with Bill Gates and Elon Musk), thank you for creating it!