December 2014 – AI Impacts

Cases of Discontinuous Technological Progress

We know of ten events which produced a robust discontinuity in progress equivalent to more than one hundred years at previous rates in some interesting metric. We know of 53 other events which produced smaller

Effect of nuclear weapons on historic trends in explosives

Nuclear weapons constituted a ~7 thousand year discontinuity in relative effectiveness factor (TNT equivalent per kg of explosive). Nuclear weapons do not appear to have clearly represented progress in the cost-effectiveness of explosives, though the

AGI-09 Survey

Baum et al. surveyed 21 attendees of the AGI-09 conference, on AGI timelines with and without extra funding. They also asked about other details of AGI development such as social impacts, and promising approaches. Their findings include the following:

Bainbridge Survey

A survey of twenty-six technology experts in 2005 produced a median of 2085 as the year in which artificial intelligence would be able to functionally replicate a human brain (p344). They rated this application 5.6/10 in beneficialness to

AI@50 Survey

A seemingly informal seven-question poll was taken of participants at the AI@50 conference in 2006. 41% of respondents said it would take more than 50 years for AI to simulate every aspect of human intelligence, and 41%

Early Views of AI

This is an incomplete list of early works we have found discussing AI or AI related problems. List 1. Claude Shannon (1950), in Programming a Computer for Playing Chess, offers the following list of “possible developments in

FHI Winter Intelligence Survey

The Future of Humanity Institute administered a survey in 2011 at their Winter Intelligence AGI impacts conference. Participants’ median estimate for a 50% chance of human-level AI was 2050. Details AI timelines question The survey included

Hanson AI Expert Survey

In a small informal survey running since 2012, AI researchers generally estimated that their subfields have moved less than ten percent of the way to human-level intelligence. Only one (in the slowest moving subfield) observed acceleration.

Müller and Bostrom AI Progress Poll

Vincent Müller and Nick Bostrom of FHI conducted a poll of four groups of AI experts in 2012-13. Combined, the median date by which they gave a 10% chance of human-level AI was 2022, and the

Kruel AI Interviews

Alexander Kruel interviewed 37 experts on areas related to AI, starting in 2011 and probably ending in 2012. Of those answering the question in a full quantitative way, median estimates for human-level AI (assuming business as usual)