PaperlayerView exact revision on arXiv

Source-linked AI summary

When Will AI Exceed Human Performance? Evidence from AI Experts

Katja Grace, John Salvatier, Allan Dafoe, Baobao Zhang, Owain Evans

arXiv:1705.08807v3cs.AIcs.CY

TL;DR

The paper addresses the need for better forecasts of transformative AI and surveys machine-learning researchers about AI timelines and consequences. Using responses from researchers at major machine-learning conferences, it estimates timelines for capabilities, HLMI, and labor automation and elicits views about explosive progress and outcomes. Respondents placed a 50% probability on HLMI within 45 years and full labor automation 122 years from now, while judging explosive progress possible but improbable and catastrophic outcomes possible.

Abstract

from arXiv · show

Advances in artificial intelligence (AI) will transform modern life by reshaping transportation, health, science, finance, and the military. To adapt public policy, we need to better anticipate these advances. Here we report the results from a large survey of machine learning researchers on their beliefs about progress in AI. Researchers predict AI will outperform humans in many activities in the next ten years, such as translating languages (by 2024), writing high-school essays (by 2026), driving a truck (by 2027), working in retail (by 2031), writing a bestselling book (by 2049), and working as a surgeon (by 2053). Researchers believe there is a 50% chance of AI outperforming humans in all tasks in 45 years and of automating all human jobs in 120 years, with Asian respondents expecting these dates much sooner than North Americans. These results will inform discussion amongst researchers and policymakers about anticipating and managing trends in AI.

Introduction

The paper surveys machine-learning researchers about AI progress, its timing, and its potential consequences to improve forecasting for policy and societal preparation. Respondents forecast rapid advances in specific capabilities, while assigning later timelines to HLMI and full labor automation, with substantial regional variation.

Motivation

  • AI advances could create major employment, infrastructure, cybersecurity, legal, military, and marketing challenges, making accurate forecasting valuable for preparation.The paper frames forecasting as relevant to both policymakers and AI developers.Source: Introduction · p. 1

Survey scope

  • The survey sampled researchers who published at NIPS or ICML and received 352 responses, covering AI capabilities, occupations, automation, and social impacts.These conferences are described as premier venues for peer-reviewed machine-learning research.Source: Survey Method · p. 1, Introduction · p. 1

Forecasts

  • 50% of respondents’ aggregate probability placed HLMI within 45 years, while full labor automation received a 50% probability 122 years from now.HLMI was defined as unaided machines accomplishing every task better and more cheaply than human workers; full labor automation used an equivalent occupational criterion.Source: Time Until Machines Outperform Humans · pp. 1–2

Intelligence explosion

  • Researchers assigned a 10% median probability to AI becoming vastly better than humans in all tasks two years after HLMI, versus 20% for explosive global technological improvement.Both estimates were considered possible but improbable, with interquartile ranges of 1–25% and 5–50%, respectively.Source: Intelligence Explosion, Outcomes, AI… · p. 4

Outcomes and safety

  • Median probabilities were 25% for a good long-run outcome, 20% for an extremely good outcome, 10% for a bad outcome, and 5% for an extremely bad outcome such as human extinction.The survey used a five-point scale for possible long-run effects of HLMI on humanity.Source: Intelligence Explosion, Outcomes, AI… · p. 4

Variation and implications

Supplementary Information

The survey covered AI timelines, intelligence-explosion probabilities, welfare implications, research inputs, and narrow AI milestones.

  • HLMI questions used three framings: direct HLMI timing, automation of all human occupations, and extrapolation from recent AI progress.Source: Survey Content · p. 7
  • Researchers were asked about the probability of an “intelligence explosion.”Source: Survey Content · p. 7
  • One question assessed the welfare implications of HLMI.Source: Survey Content · p. 7
  • The survey examined how inputs such as hardware progress affect the rate of AI research.Source: Survey Content · p. 7

5. Two questions about sources of disagreement about AI timelines and “AI Safety.”

This section defines HLMI, describes how predictions were elicited and aggregated, and records additional survey coverage of AI milestones, safety, and demographics.

Survey Content

  • The survey also included 32 narrow AI milestone questions, AI Safety questions, and demographic questions administered online.Source: §5 · p. 7
  • HLMI means a machine that is better than humans at all tasks.Source: §5 · p. 7

Respondent Data

Elicitation of Beliefs

  • Timeline questions used fixed-probability and fixed-years framings, which sampled different CDF points and produced framing effects among respondents.Source: Elicitation of Beliefs · p. 7

Analysis

  • Aggregate distributions fitted a gamma CDF to each respondent, combined individual fits into a mixture distribution, and used bootstrapping for 95% confidence intervals.Source: Statistics · p. 8

Supplementary Figures

The supplementary figures show aggregate HLMI forecasts by demographic group and compare two question framings with their averaged forecast.

  • Figure S1 groups aggregate HLMI forecast CDFs by country, seniority quartile, and citation-count quartile.Source: Figure S1 · p. 9
  • Figure S2 compares aggregate forecasts from fixed-probability and fixed-years framings and displays their combined average.Source: Figure S2 · p. 10

Supplementary Tables

The supplementary material provides implementation details, respondent comparisons, survey-response summaries, and operational descriptions of AI milestones.

Automation Predictions

HLMI Regression

Respondent Comparison

AI Milestones

  • The milestone descriptions define capabilities spanning novel-category learning, one-shot learning, video generation, speech transcription, and text-to-speech.Source: S5: Description of AI Milestones · p. 15
  • Additional milestones cover mathematical research, Putnam-level mathematics, Go, StarCraft, random-game play, Atari, laundry folding, city racing, LEGO assembly, sorting, and code generation.Source: S5: Description of AI Milestones · p. 16

Related papers

Loading 1705.08807v3…