|
|
|
|
— |
ai_timelines:predictions_of_human-level_ai_timelines:ai_timeline_surveys:2016_espai_questions_printout [2022/09/21 07:37] (current)
|
| | ====== 2016 ESPAI questions printout ====== |
| | |
| | // Published 26 June, 2017; last updated 28 May, 2020 // |
| | |
| | <HTML> |
| | <p>This is a list of questions from the <a href="/doku.php?id=ai_timelines:predictions_of_human-level_ai_timelines:ai_timeline_surveys:2016_expert_survey_on_progress_in_ai">2016 Expert Survey on Progress in AI</a>.</p> |
| | </HTML> |
| | |
| | |
| | ===== Details ===== |
| | |
| | |
| | <HTML> |
| | <p>This page is a printout of questions from the <a href="/doku.php?id=ai_timelines:predictions_of_human-level_ai_timelines:ai_timeline_surveys:2016_expert_survey_on_progress_in_ai">2016 Expert Survey on Progress in AI</a> provided by the Qualtrics website as a word document, and then copied here, for searchability. It contains formatting differences with the survey as received by participants, and probably typographic errors, due to the importing process. It also contains only parts of the randomization logic, while missing other parts. The survey questions are available as a pdf <a href="https://www.dropbox.com/s/99os4grxlhf744m/Final%202016%20Expert%20Survey%20on%20Progress%20in%20AI.pdf?dl=0">here</a>.</p> |
| | </HTML> |
| | |
| | |
| | ===== Printout ===== |
| | |
| | |
| | <HTML> |
| | <p>16-05-17 AI Survey 12 – final</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>consent_tim Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>consent 2016 Expert Survey on Progress in AI Welcome. We are conducting a study of progress in artificial intelligence and are interested in your understanding of developments in the field. Our estimated median time for completing this survey is 12 minutes. Your responses will be kept confidential. Many of the questions involve substantial uncertainties. Please just give us your current best guesses. There are no known risks associated with this study. Although this study may not benefit you personally, we hope that our results will add to the knowledge about progress in AI technology. If you have questions about your rights as a research participant, you may contact the Yale University Human Subjects Committee: 203-785-4688, human.subjects@yale.edu. Additional information is available at: http://www.yale.edu/hrpp/participants/index.html Participation in this study is completely voluntary. You are free to decline to participate and to end participation at any time for any reason. By continuing to the next page, you agree to participate in the survey.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hb_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hb_def 1 of 7 The following questions ask about ‘high–level machine intelligence’ (HLMI). Say we have ‘high-level machine intelligence’ when unaided machines can accomplish every task better and more cheaply than human workers. Ignore aspects of tasks for which being a human is intrinsically advantageous, e.g. being accepted as a jury member. Think feasibility, not adoption.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If fixedprobabilities Is Equal to 1</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hb_a For the purposes of this question, assume that human scientific activity continues without major negative disruption. How many years until you expect:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <table> |
| | <tbody> |
| | <tr> |
| | <td width="239"></td> |
| | <td width="239"> (1)</td> |
| | </tr> |
| | <tr> |
| | <td width="239">a 10% probability of HLMI existing? (1)</td> |
| | <td width="239"></td> |
| | </tr> |
| | <tr> |
| | <td width="239">a 50% probability of HLMI existing? (2)</td> |
| | <td width="239"></td> |
| | </tr> |
| | <tr> |
| | <td width="239">a 90% probability of HLMI existing? (3)</td> |
| | <td width="239"></td> |
| | </tr> |
| | </tbody> |
| | </table> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If fixedprobabilities Is Equal to 0</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hb_b For the purposes of this question, assume that human scientific activity continues without major negative disruption. How likely is it that HLMI exists:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>in 10 years? (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>in 20 years? (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>in 40 years? (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 20</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hb_comment Do you have any comments on your interpretation of this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 40</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hb_consider Which considerations were important in your answers to this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_a_jobs 1 of 7 Say an occupation becomes fully automatable when unaided machines can accomplish it better and more cheaply than human workers. Ignore aspects of occupations for which being a human is intrinsically advantageous, e.g. being accepted as a jury member. Think feasibility, not adoption. We want to know how many years you think will pass before the following present-day occupations will be fully automatable. Please tell us your best guess of when you think there will be a small chance (10% chance), a roughly even chance (50% chance), and a high chance (90% chance).</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <table> |
| | <tbody> |
| | <tr> |
| | <td width="120"></td> |
| | <td width="120">Years until small chance (10%) (1)</td> |
| | <td width="120">Years until even chance (50%) (2)</td> |
| | <td width="120">Years until high chance (90%) (3)</td> |
| | </tr> |
| | <tr> |
| | <td width="120">Truck driver (hj_a_jobs_1)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | <tr> |
| | <td width="120">Surgeon (hj_a_jobs_2)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | <tr> |
| | <td width="120">Retail salesperson (hj_a_jobs_3)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | <tr> |
| | <td width="120">AI researcher (hj_a_jobs_4)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | </tbody> |
| | </table> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_a_final What is an existing human occupation that you think will be among the final ones to be fully automatable? Remember to consider feasibility, not adoption.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_a_final_pred How many years do you expect to pass before you think there is a small/even/high chance that this occupation will be fully automatable?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>High chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_a_full Say we have reached ‘full automation of labor’ when all occupations are fully automatable. That is, when for any occupation, machines could be built to carry out the task better and more cheaply than human workers.In how many years do you expect full automation of labor, with small/even/high chance?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>High chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 20</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_a_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 40</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_a_consider Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_jobs 1 of 7 Say an occupation becomes fully automatable when unaided machines can accomplish it better and more cheaply than human workers. Ignore aspects of occupations for which being a human is intrinsically advantageous, e.g. being accepted as a jury member. Think feasibility, not adoption. We want to know how likely you think it is that the following present-day occupations will be fully automatable at future dates. Please tell us your best guess of the chance that they will be fully automatable within the next 10 years, within the next 20 years, and within the next 50 years.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <table> |
| | <tbody> |
| | <tr> |
| | <td width="120"></td> |
| | <td width="120">% chance in 10 years (1)</td> |
| | <td width="120">% chance in 20 years (2)</td> |
| | <td width="120">% chance in 50 years (3)</td> |
| | </tr> |
| | <tr> |
| | <td width="120">Truck driver (hj_b_jobs_1)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | <tr> |
| | <td width="120">Surgeon (hj_b_jobs_2)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | <tr> |
| | <td width="120">Retail salesperson (hj_b_jobs_3)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | <tr> |
| | <td width="120">AI researcher (hj_b_jobs_4)</td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | <td width="120"></td> |
| | </tr> |
| | </tbody> |
| | </table> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_final What is an existing human occupation that you think will be among the final ones to be fully automatable? Remember to consider feasibility, not adoption.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_final_pred How likely do you think it is that this occupation will be fully automatable within the next 10/20/50 years?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_full Say we have reached ‘full automation of labor’ when all occupations are fully automatable. That is, when for any occupation, machines could be built to carry out the task better and more cheaply than human workers. How likely do you think it is that full automation of labor will happen within the next 10/20/50 years?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 20</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 40</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hj_b_consier Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_time_def 2 of 7 The following questions ask about ‘high–level machine intelligence’ (HLMI). Say we have ‘high-level machine intelligence’ when unaided machines can accomplish every task better and more cheaply than human workers. Ignore aspects of tasks for which being a human is intrinsically advantageous, e.g. being accepted as a jury member. Think feasibility, not adoption.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_1 Assume that HLMI will exist at some point. How likely do you then think it is that the rate of global technological improvement will dramatically increase (e.g. by a factor of ten) as a result of machine intelligence:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Within two years of that point? (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Within thirty years of that point? (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_2 Assume that HLMI will exist at some point. How likely do you think it is that there will be machine intelligence that is vastly better than humans at all professions (i.e. that is vastly more capable or vastly cheaper):</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Within two years of that point? (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Within thirty years of that point? (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_3 Some people have argued the following: If AI systems do nearly all research and development, improvements in AI will accelerate the pace of technological progress, including further progress in AI. Over a short period (less than 5 years), this feedback loop could cause technological progress to become more than an order of magnitude faster. How likely do you find this argument to be broadly correct?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Quite unlikely (0-20%) (5)</div></li> |
| | <li><div class="li">Unlikely (21-40%) (4)</div></li> |
| | <li><div class="li">About even chance (41-60%) (3)</div></li> |
| | <li><div class="li">Likely (61-80%) (2)</div></li> |
| | <li><div class="li">Quite likely (81-100%) (1)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 40</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 50</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 50</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 60</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ie_consider Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>vb_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>vb_def 3 of 7 The following questions ask about ‘high–level machine intelligence’ (HLMI). Say we have ‘high-level machine intelligence’ when unaided machines can accomplish every task better and more cheaply than human workers. Ignore aspects of tasks for which being a human is intrinsically advantageous, e.g. being accepted as a jury member. Think feasibility, not adoption.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>vb_1 Assume for the purpose of this question that HLMI will at some point exist. How positive or negative do you expect the overall impact of this to be on humanity, in the long run? Please answer by saying how probable you find the following kinds of impact, with probabilities adding to 100%:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>______ Extremely good (e.g. rapid growth in human flourishing) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>______ On balance good (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>______ More or less neutral (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>______ On balance bad (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>______ Extremely bad (e.g. human extinction) (5)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 60</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 70</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>vb_comment Do you have any comments on your interpretation of this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 70</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 80</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>vb_consider Which considerations were important in your answers to this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_def 4 of 7 The next questions ask about the sensitivity of progress in AI capabilities to changes in inputs. ‘Progress in AI capabilities’ is an imprecise concept, so we are asking about progress as you naturally conceive of it, and looking for approximate answers.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_1 Imagine that over the past decade, only half as much researcher effort had gone into AI research. For instance, if there were actually 1,000 researchers, imagine that there had been only 500 researchers (of the same quality).How much less progress in AI capabilities would you expect to have seen? e.g. If you think progress is linear in the number of researchers, so 50% less progress would have been made, write ’50’. If you think only 20% less progress would have been made write ’20’.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_2 Over the last 10 years the cost of computing hardware has fallen by a factor of 20. Imagine instead that the cost of computing hardware had fallen by only a factor of 5 over that time (around half as far on a log scale). How much less progress in AI capabilities would you expect to have seen? e.g. If you think progress is linear in 1/cost, so that 1-5/20=75% less progress would have been made, write ’75’. If you think only 20% less progress would have been made write ’20’.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_3 Imagine that over the past decade, there had only been half as much effort put into increasing the size and availability of training datasets. For instance, perhaps there are only half as many datasets, or perhaps existing datasets are substantially smaller or lower quality.How much less progress in AI capabilities would you expect to have seen? e.g. If you think 20% less progress would have been made, write ‘20’</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_4 Imagine that over the past decade, AI research had half as much funding (in both academic and industry labs). For instance, if the average lab had a budget of $20 million each year, suppose their budget had only been $10 million each year. How much less progress in AI capabilities would you expect to have seen? e.g. If you think 20% less progress would have been made, write ‘20’</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_5 Imagine that over the past decade, there had been half as much progress in AI algorithms. You might imagine this as conceptual insights being half as frequent. How much less progress in AI capabilities would you expect to have seen? e.g. If you think 20% less progress would have been made, write ‘20’</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 80</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 100</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>c_consider Which considerations were important in your answers to these question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_area 4 of 7 Which AI research area have you worked in for the longest time?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_howlong How long have you worked in this area?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_1 Consider three levels of progress or advancement in this area: A. Where the area was when you started working in it B. Where it is now C. Where it would need to be for AI software to have roughly human level abilities at the tasks studied in this area What fraction of the distance between where progress was when you started working in the area (A) and where it would need to be to attain human level abilities in the area (C) have we come so far (B)?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_2 Divide the period you have worked in the area into two halves: the first and the second. In which half was the rate of progress in your area higher?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">The first half (1)</div></li> |
| | <li><div class="li">The second half (2)</div></li> |
| | <li><div class="li">They were about the same (3)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 20</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 30</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 40</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>hh_consider Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_1 4 of 7 To what extent do you think you disagree with the typical AI researcher about when HLMI will exist?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">A lot (17)</div></li> |
| | <li><div class="li">A moderate amount (18)</div></li> |
| | <li><div class="li">Not much (19)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_2 If you disagree, why do you think that is?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_3 To what extent do you think people’s concerns about future risks from AI are due to misunderstandings of AI research?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Almost entirely (1)</div></li> |
| | <li><div class="li">To a large extent (2)</div></li> |
| | <li><div class="li">Somewhat (4)</div></li> |
| | <li><div class="li">Not much (3)</div></li> |
| | <li><div class="li">Hardly at all (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_4 What do you think are the most important misunderstandings, if there are any?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 80</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 100</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ms_consider Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_def 5 of 7 How many years until you think the following AI tasks will be feasible with: a small chance (10%)? an even chance (50%)? a high chance (90%)? Let a task be ‘feasible’ if one of the best resourced labs could implement it in less than a year if they chose to. Ignore the question of whether they would choose to. Tasks</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_1 Translate a text written in a newly discovered language into English as well as a team of human experts, using a single other document in both languages (like a Rosetta stone). Suppose all of the words in the text can be found in the translated document, and that the language is a difficult one.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_2 Translate speech in a new language given only unlimited films with subtitles in the new language. Suppose the system has access to training data for other languages, of the kind used now (e.g. same text in two languages for many languages and films with subtitles in many languages).</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_3 Perform translation about as good as a human who is fluent in both languages but unskilled at translation, for most types of text, and for most popular languages (including languages that are known to be difficult, like Czech, Chinese and Arabic).</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_4 Provide phone banking services as well as human operators can, without annoying customers more than humans. This includes many one-off tasks, such as helping to order a replacement bank card or clarifying how to use part of the bank website to a customer.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_5 Correctly group images of previously unseen objects into classes, after training on a similar labeled dataset containing completely different classes. The classes should be similar to the ImageNet classes.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_6 One-shot learning: see only one labeled image of a new object, and then be able to recognize the object in real world scenes, to the extent that a typical human can (i.e. including in a wide variety of settings). For example, see only one image of a platypus, and then be able to recognize platypuses in nature photos. The system may train on labeled images of other objects. Currently, deep networks often need hundreds of examples in classification tasks1, but there has been work on one-shot learning for both classification2 and generative tasks3. 1 Lake et al. (2015). Building Machines That Learn and Think Like People 2 Koch (2015). Siamese Neural Networks for One-Shot Image Recognition 3 Rezende et al. (2016). One-Shot Generalization in Deep Generative Models</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_7 See a short video of a scene, and then be able to construct a 3D model of the scene good enough to create a realistic video of the same scene from a substantially different angle.For example, constructing a short video of walking through a house from a video taking a very different path through the house.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_8 Transcribe human speech with a variety of accents in a noisy environment as well as a typical human can.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_9 Take a written passage and output a recording that can’t be distinguished from a voice actor, by an expert listener.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_10 Routinely and autonomously prove mathematical theorems that are publishable in top mathematics journals today, including generating the theorems to prove.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_11 Perform as well as the best human entrants in the Putnam competition—a math contest whose questions have known solutions, but which are difficult for the best young mathematicians.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_12 Defeat the best Go players, training only on as many games as the best Go players have played. For reference, DeepMind’s AlphaGo has probably played a hundred million games of self-play, while Lee Sedol has probably played 50,000 games in his life1. 1 Lake et al. (2015). Building Machines That Learn and Think Like People</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_13 Beat the best human Starcraft 2 players at least 50% of the time, given a video of the screen. Starcraft 2 is a real time strategy game characterized by: Continuous time play Huge action space Partial observability of enemies Long term strategic play, e.g. preparing for and then hiding surprise attacks.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_14 Play a randomly selected computer game, including difficult ones, about as well as a human novice, after playing the game less than 10 minutes of game time. The system may train on other games.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_15 Play new levels of Angry Birds better than the best human players. Angry Birds is a game where players try to efficiently destroy 2D block towers with a catapult. For context, this is the goal of the IJCAI Angry Birds AI competition1. 1 aibirds.org</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_16 Outperform professional game testers on all Atari games using no game-specific knowledge. This includes games like Frostbite, which require planning to achieve sub-goals and have posed problems for deep Q-networks1, 2. 1 Mnih et al. (2015). Human-level control through deep reinforcement learning 2 Lake et al. (2015). Building Machines That Learn and Think Like People</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_17 Outperform human novices on 50% of Atari games after only 20 minutes of training play time and no game specific knowledge. For context, the original Atari playing deep Q-network outperforms professional game testers on 47% of games1, but used hundreds of hours of play to train2. 1 Mnih et al. (2015). Human-level control through deep reinforcement learning 2 Lake et al. (2015). Building Machines That Learn and Think Like People</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_18 Fold laundry as well and as fast as the median human clothing store employee.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_19 Beat the fastest human runners in a 5 kilometer race through city streets using a bipedal robot body.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_20 Physically assemble any LEGO set given the pieces and instructions, using non-specialized robotics hardware. For context, Fu 20161 successfully joins single large LEGO pieces using model based reinforcement learning and online adaptation. 1 Fu et al. (2016). One-Shot Learning of Manipulation Skills with Online Dynamics Adaptation and Neural Network Priors</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_21 Learn to efficiently sort lists of numbers much larger than in any training set used, the way Neural GPUs can do for addition1, but without being given the form of the solution. For context, Neural Turing Machines have not been able to do this2, but Neural Programmer-Interpreters3 have been able to do this by training on stack traces (which contain a lot of information about the form of the solution). 1 Kaiser & Sutskever (2015). Neural GPUs Learn Algorithms 2 Zaremba & Sutskever (2015). Reinforcement Learning Neural Turing Machines 3 Reed & de Freitas (2015). Neural Programmer-Interpreters</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_22 Write concise, efficient, human-readable Python code to implement simple algorithms like quicksort. That is, the system should write code that sorts a list, rather than just being able to sort lists. Suppose the system is given only: A specification of what counts as a sorted list Several examples of lists undergoing sorting by quicksort</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_23 Answer any “easily Googleable” factoid questions posed in natural language better than an expert on the relevant topic (with internet access), having found the answers on the internet. Examples of factoid questions: “What is the poisonous substance in Oleander plants?” “How many species of lizard can be found in Great Britain?”</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_24 Answer any “easily Googleable” factual but open ended question posed in natural language better than an expert on the relevant topic (with internet access), having found the answers on the internet. Examples of open ended questions: “What does it mean if my lights dim when I turn on the microwave?” “When does home insurance cover roof replacement?”</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_25 Give good answers in natural language to factual questions posed in natural language for which there are no definite correct answers. For example:”What causes the demographic transition?”, “Is the thylacine extinct?”, “How safe is seeing a chiropractor?”</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_26 Write an essay for a high-school history class that would receive high grades and pass plagiarism detectors. For example answer a question like ‘How did the whaling industry affect the industrial revolution?’</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_27 Compose a song that is good enough to reach the US Top 40. The system should output the complete song as an audio file.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_28 Produce a song that is indistinguishable from a new song by a particular artist, e.g. a song that experienced listeners can’t distinguish from a new song by Taylor Swift.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_29 Write a novel or short story good enough to make it to the New York Times best-seller list.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_30 For any computer game that can be played well by a machine, explain the machine’s choice of moves in a way that feels concise and complete to a layman.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_31 Play poker well enough to win the World Series of Poker.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_32 After spending time in a virtual world, output the differential equations governing that world in symbolic form.For example, the agent is placed in a game engine where Newtonian mechanics holds exactly and the agent is then able to conduct experiments with a ball and output Newton’s laws of motion.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>small chance (10%) (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>even chance (50%) (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>high chance (90%) (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 0</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 10</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 10</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 20</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>ta_consider Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_def 5 of 7 How likely do you think it is that the following AI tasks will be feasible within the next: 10 years? 20 years? 50 years? Let a task be ‘feasible’ if one of the best resourced labs could implement it in less than a year if they chose to. Ignore the question of whether they would choose to. Tasks</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_1 Translate a text written in a newly discovered language into English as well as a team of human experts, using a single other document in both languages (like a Rosetta stone). Suppose all of the words in the text can be found in the translated document, and that the language is a difficult one.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_2 Translate speech in a new language given only unlimited films with subtitles in the new language. Suppose the system has access to training data for other languages, of the kind used now (e.g. same text in two languages for many languages and films with subtitles in many languages).</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (5)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (6)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_3 Perform translation about as good as a human who is fluent in both languages but unskilled at translation, for most types of text, and for most popular languages (including languages that are known to be difficult, like Czech, Chinese and Arabic).</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_4 Provide phone banking services as well as human operators can, without annoying customers more than humans. This includes many one-off tasks, such as helping to order a replacement bank card or clarifying how to use part of the bank website to a customer.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_5 Correctly group images of previously unseen objects into classes, after training on a similar labeled dataset containing completely different classes. The classes should be similar to the ImageNet classes.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_6 One-shot learning: see only one labeled image of a new object, and then be able to recognize the object in real world scenes, to the extent that a typical human can (i.e. including in a wide variety of settings). For example, see only one image of a platypus, and then be able to recognize platypuses in nature photos. The system may train on labeled images of other objects. Currently, deep networks often need hundreds of examples in classification tasks1, but there has been work on one-shot learning for both classification2 and generative tasks3. 1 Lake et al. (2015). Building Machines That Learn and Think Like People 2 Koch (2015). Siamese Neural Networks for One-Shot Image Recognition 3 Rezende et al. (2016). One-Shot Generalization in Deep Generative Models</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_7 See a short video of a scene, and then be able to construct a 3D model of the scene that is good enough to create a realistic video of the same scene from a substantially different angle.For example, constructing a short video of walking through a house from a video taking a very different path through the house.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_8 Transcribe human speech with a variety of accents in a noisy environment as well as a typical human can.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_9 Take a written passage and output a recording that can’t be distinguished from a voice actor, by an expert listener.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_10 Routinely and autonomously prove mathematical theorems that are publishable in top mathematics journals today, including generating the theorems to prove.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_11 Perform as well as the best human entrants in the Putnam competition—a math contest whose questions have known solutions, but which are difficult for the best young mathematicians.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_12 Defeat the best Go players, training only on as many games as the best Go players have played. For reference, DeepMind’s AlphaGo has probably played a hundred million games of self-play, while Lee Sedol has probably played 50,000 games in his life1. 1 Lake et al. (2015). Building Machines That Learn and Think Like People</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_13 Beat the best human Starcraft 2 players at least 50% of the time, given a video of the screen. Starcraft 2 is a real time strategy game characterized by: Continuous time play Huge action space Partial observability of enemies Long term strategic play, e.g. preparing for and then hiding surprise attacks.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_14 Play a randomly selected computer game, including difficult ones, about as well as a human novice, after playing the game less than 10 minutes of game time. The system may train on other games.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_15 Play new levels of Angry Birds better than the best human players. Angry Birds is a game where players try to efficiently destroy 2D block towers with a catapult. For context, this is the goal of the IJCAI Angry Birds AI competition1. 1 aibirds.org</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_16 Outperform professional game testers on all Atari games using no game-specific knowledge. This includes games like Frostbite, which require planning to achieve sub-goals and have posed problems for deep Q-networks1, 2. 1 Mnih et al. (2015). Human-level control through deep reinforcement learning 2 Lake et al. (2015). Building Machines That Learn and Think Like People</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_17 Outperform human novices on 50% of Atari games after only 20 minutes of training play time and no game specific knowledge. For context, the original Atari playing deep Q-network outperforms professional game testers on 47% of games1, but used hundreds of hours of play to train2. 1 Mnih et al. (2015). Human-level control through deep reinforcement learning 2 Lake et al. (2015). Building Machines That Learn and Think Like People</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_18 Fold laundry as well and as fast as the median human clothing store employee.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_19 Beat the fastest human runners in a 5 kilometer race through city streets using a bipedal robot body.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_20 Physically assemble any LEGO set given the pieces and instructions, using non-specialized robotics hardware. For context, Fu 20161 successfully joins single large LEGO pieces using model based reinforcement learning and online adaptation. 1 Fu et al. (2016). One-Shot Learning of Manipulation Skills with Online Dynamics Adaptation and Neural Network Priors</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_21 Learn to efficiently sort lists of numbers much larger than in any training set used, the way Neural GPUs can do for addition1, but without being given the form of the solution. For context, Neural Turing Machines have not been able to do this2, but Neural Programmer-Interpreters3 have been able to do this by training on stack traces (which contain a lot of information about the form of the solution). 1 Kaiser & Sutskever (2015). Neural GPUs Learn Algorithms 2 Zaremba & Sutskever (2015). Reinforcement Learning Neural Turing Machines 3 Reed & de Freitas (2015). Neural Programmer-Interpreters</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_22 Write concise, efficient, human-readable Python code to implement simple algorithms like quicksort. That is, the system should write code that sorts a list, rather than just being able to sort lists. Suppose the system is given only: A specification of what counts as a sorted list Several examples of lists undergoing sorting by quicksort</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_23 Answer any “easily Googleable” factoid questions posed in natural language better than an expert on the relevant topic (with internet access), having found the answers on the internet. Examples of factoid questions: “What is the poisonous substance in Oleander plants?” “How many species of lizard can be found in Great Britain?”</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_24 Answer any “easily Googleable” factual but open ended question posed in natural language better than an expert on the relevant topic (with internet access), having found the answers on the internet. Examples of open ended questions: “What does it mean if my lights dim when I turn on the microwave?” “When does home insurance cover roof replacement?”</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_25 Give good answers in natural language to factual questions posed in natural language for which there are no definite correct answers. For example:”What causes the demographic transition?”, “Is the thylacine extinct?”, “How safe is seeing a chiropractor?”</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_26 Write an essay for a high-school history class that would receive high grades and pass plagiarism detectors. For example answer a question like ‘How did the whaling industry affect the industrial revolution?’</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_27 Compose a song that is good enough to reach the US Top 40. The system should output the complete song as an audio file.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_28 Produce a song that is indistinguishable from a new song by a particular artist, e.g. a song that experienced listeners can’t distinguish from a new song by Taylor Swift.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_29 Write a novel or short story good enough to make it to the New York Times best-seller list.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_30 For any computer game that can be played well by a machine, explain the machine’s choice of moves in a way that feels concise and complete to a layman.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_31 Play poker well enough to win the World Series of Poker.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_32 After spending time in a virtual world, output the differential equations governing that world in symbolic form.For example, the agent is placed in a game engine where Newtonian mechanics holds exactly and the agent is then able to conduct experiments with a ball and output Newton’s laws of motion.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>10 years (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>20 years (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>50 years (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 0</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 10</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_comment Do you have any comments on your interpretation of these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 10</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than 20</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>tb_consider Which considerations were important in your answers to these questions? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sq_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sq_def 6 of 7 Stuart Russell summarizes an argument for why highly advanced AI might pose a risk as follows: The primary concern [with highly advanced AI] is not spooky emergent consciousness but simply the ability to make high-quality decisions. Here, quality refers to the expected outcome utility of actions taken […]. Now we have a problem: 1. The utility function may not be perfectly aligned with the values of the human race, which are (at best) very difficult to pin down. 2. Any sufficiently capable intelligent system will prefer to ensure its own continued existence and to acquire physical and computational resources – not for their own sake, but to succeed in its assigned task. A system that is optimizing a function of n variables, where the objective depends on a subset of size k<n, will often set the remaining unconstrained variables to extreme values; if one of those unconstrained variables is actually something we care about, the solution found may be highly undesirable. This is essentially the old story of the genie in the lamp, or the sorcerer’s apprentice, or King Midas: you get exactly what you ask for, not what you want.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sq_1 Do you think this argument points at an important problem?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">No, not a real problem. (1)</div></li> |
| | <li><div class="li">No, not an important problem. (2)</div></li> |
| | <li><div class="li">Yes, a moderately important problem. (3)</div></li> |
| | <li><div class="li">Yes, a very important problem. (5)</div></li> |
| | <li><div class="li">Yes, among the most important problems in the field. (4)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>aq_2 How valuable is it to work on this problem today, compared to other problems in AI?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Much less valuable (1)</div></li> |
| | <li><div class="li">Less valuable (2)</div></li> |
| | <li><div class="li">As valuable as other problems (3)</div></li> |
| | <li><div class="li">More valuable (4)</div></li> |
| | <li><div class="li">Much more valuable (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sq_3 How hard do you think this problem is compared to other problems in AI?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Much easier (1)</div></li> |
| | <li><div class="li">Easier (2)</div></li> |
| | <li><div class="li">As hard as other problems (3)</div></li> |
| | <li><div class="li">Harder (4)</div></li> |
| | <li><div class="li">Much harder (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sq_comment Do you have any comments on your interpretation of this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 100</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sq_consider Which considerations were important in your answers to this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sr_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sr_def 6 of 7 Let ‘AI safety research’ include any AI-related research that, rather than being primarily aimed at improving the capabilities of AI systems, is instead primarily aimed at minimizing potential risks of AI systems (beyond what is already accomplished for those goals by increasing AI system capabilities). Examples of AI safety research might include: Improving the human-interpretability of machine learning algorithms for the purpose of improving the safety and robustness of AI systems, not focused on improving AI capabilities Research on long-term existential risks from AI systems AI-specific formal verification research Policy research about how to maximize the public benefits of AI</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sr_1 How much should society prioritize AI safety research, relative to how much it is currently prioritized?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Much less (1)</div></li> |
| | <li><div class="li">Less (2)</div></li> |
| | <li><div class="li">About the same (3)</div></li> |
| | <li><div class="li">More (4)</div></li> |
| | <li><div class="li">Much more (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 80</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sr_comment Do you have any comments on your interpretation of this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Display This Question:</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>If random Is Greater Than 90</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>And random Is Less Than or Equal to 100</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>sr_consider Which considerations were important in your answer to this question? (optional)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>dem_time Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>dem_1 7 of 7 How much thought have you given in the past to when HLMI (or something similar) will be developed?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Very little. e.g. “I can’t remember thinking about this.” (1)</div></li> |
| | <li><div class="li">A little. e.g. “It has come up in conversation a few times” (2)</div></li> |
| | <li><div class="li">A moderate amount. e.g. “I read something about it now and again” (3)</div></li> |
| | <li><div class="li">A lot. e.g. “I have thought enough to have my own views on the topic” (4)</div></li> |
| | <li><div class="li">A great deal. e.g. “This has been a particular interest of mine” (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>dem_2 How much thought have you given in the past to social impacts of smarter-than-human machines?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Very little. e.g. “I can’t remember thinking about this.” (1)</div></li> |
| | <li><div class="li">A little. e.g. “It has come up in conversation a few times” (2)</div></li> |
| | <li><div class="li">A moderate amount. e.g. “I read something about it now and again” (3)</div></li> |
| | <li><div class="li">A lot. e.g. “I have thought enough to have my own views on the topic” (4)</div></li> |
| | <li><div class="li">A great deal. e.g. “This has been a particular interest of mine” (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>dem_3 Are you an AI researcher?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Yes (4)</div></li> |
| | <li><div class="li">No (5)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>dem_4 What are your main areas of research?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>dem_5 Where do you work?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Industry (1)</div></li> |
| | <li><div class="li">Academia (2)</div></li> |
| | <li><div class="li">Other (3)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Q203 Timing</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>First Click (1)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Last Click (2)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Page Submit (3)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>Click Count (4)</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>end Thank you for contributing to the Expert Survey on Progress in AI! We will send you output from this research as it becomes available. We are sending every 10th person $250 as an expression of gratitude for completing the survey. We are a group of researchers interested in measuring and understanding AI progress and its implications. If you are interested in learning more about this research, please click here or email us. Katja Grace, Machine Intelligence Research Institute John Salvatier, Machine Intelligence Research Institute Allan Dafoe, Yale University Baobao Zhang, Massachusetts Institute of Technology</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>print Would you like to be recognized in print as an expert participant in this survey?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Yes (1)</div></li> |
| | <li><div class="li">No (2)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>autojobs A group from Oxford University will soon conduct a survey on Automation and Jobs. It will explore some of the topics in this survey in more depth. Would you like to receive an invitation to participate in it?</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">Yes (1)</div></li> |
| | <li><div class="li">No (2)</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>comment_box If you have any questions or comments for us, please feel free to share them below.</p> |
| | </HTML> |
| | |
| | |
| |
