arguments_for_ai_risk:quantitative_estimates_of_ai_risk
Differences
This shows you the differences between two versions of the page.
| Next revision | Previous revision | ||
|
arguments_for_ai_risk:quantitative_estimates_of_ai_risk [2023/06/14 22:37] jeffreyheninger created |
arguments_for_ai_risk:quantitative_estimates_of_ai_risk [2023/12/01 18:15] (current) harlanstewart |
||
|---|---|---|---|
| Line 1: | Line 1: | ||
| ====== Quantitative Estimates of AI Risk ====== | ====== Quantitative Estimates of AI Risk ====== | ||
| + | /* | ||
| + | COMMENT: | ||
| + | Things to add to this: | ||
| + | - https:// | ||
| + | */ | ||
| // This page is in an early draft. It is very incomplete and may contain errors. // | // This page is in an early draft. It is very incomplete and may contain errors. // | ||
| Line 10: | Line 15: | ||
| For quotes from specific prominent people working on AI, see [[arguments_for_ai_risk: | For quotes from specific prominent people working on AI, see [[arguments_for_ai_risk: | ||
| + | For expert surveys about AI risk, see [[uncategorized: | ||
| This page draws heavily from [[https:// | This page draws heavily from [[https:// | ||
| Line 15: | Line 21: | ||
| ===== Quantitative Estimates ===== | ===== Quantitative Estimates ===== | ||
| - | ==== Individuals ==== | + | The table below includes estimates from individuals working in AI Safety of how likely very bad outcomes due to AI are. |
| + | |||
| + | Many of the individuals expressed [[https:// | ||
| + | |||
| + | Each estimate includes: | ||
| + | * The person making the estimate. | ||
| + | * The year the estimate was made. | ||
| + | * What exactly is being estimated. Different people have different explanation of what "very bad" looks like, and some people have given multiple conditional estimates. | ||
| + | * The estimate the individual gives for the probability that AI development causes a very bad outcome. | ||
| + | * The source for this estimate. | ||
| + | * Whether this is the person' | ||
| + | |||
| + | The estimates are in no particular order. The table can be sorted by clicking at the top of each column. | ||
| < | < | ||
| Line 49: | Line 67: | ||
| <th onclick=" | <th onclick=" | ||
| <th onclick=" | <th onclick=" | ||
| - | <th onclick=" | + | <th onclick=" |
| <th onclick=" | <th onclick=" | ||
| + | <th onclick=" | ||
| </tr> | </tr> | ||
| <tr> | <tr> | ||
| - | <td>Toby Ord</ | + | <td>Katja Grace</ |
| - | <td>2020</ | + | <td>2023</ |
| - | <td>Existential catastrophe by 2120 as a result of unaligned | + | <td>Bad future because |
| - | < | + | < |
| - | < | + | < |
| + | < | ||
| </tr> | </tr> | ||
| <tr> | <tr> | ||
| - | <td>Joe Carlsmith</ | + | <td>Joseph |
| < | < | ||
| < | < | ||
| < | < | ||
| < | < | ||
| + | < | ||
| </tr> | </tr> | ||
| <tr> | <tr> | ||
| - | <td>Katja Grace</ | + | <td>Joseph Carlsmith</ |
| - | <td>2023</ | + | <td>2022</ |
| - | <td>Bad future because | + | <td>Existential catastrophe by 2070 from advanced, planning, strategic |
| - | < | + | < |
| - | < | + | <td>Update to: <a href=" |
| + | < | ||
| </tr> | </tr> | ||
| <tr> | <tr> | ||
| Line 79: | Line 101: | ||
| < | < | ||
| < | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| </tr> | </tr> | ||
| <tr> | <tr> | ||
| Line 86: | Line 125: | ||
| < | < | ||
| < | < | ||
| + | < | ||
| </tr> | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | <tr> | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | < | ||
| + | </tr> | ||
| + | |||
| </ | </ | ||
| Line 150: | Line 335: | ||
| </ | </ | ||
| - | ==== Surveys | + | ===== Framings |
| + | |||
| + | Different people use different framings to arrive at their estimate of AI risk. The most common framing seems to be to describe a model of what the risk from advanced AI looks like, assign probabilities to various components of that model, and then calculate the existential risk from AI on the basis of this model. Another framing is to describe various scenarios for the future of AI, assign probabilities to the various scenarios, and then add together the probabilities of the different scenarios to determine the total existential risk from AI. There are also some people who give a probability without describing what framing they used to get this number. | ||
| + | |||
| + | Below is an example of each of these two framings, due to Joseph Carlsmith and Peter Wildeford, respectively. Both individuals have updated their estimates since publishing their framing, so neither probability breakdown reflects the author' | ||
| + | |||
| + | ==== Model ==== | ||
| + | |||
| + | One example of using a model to calculate the existential risk from AI is due to Joseph Carlsmith. He calculates AI-risk by 2070 by breaking it down in the following way: | ||
| + | |||
| + | - It will become possible and financially feasible to build APS [advanced, planning, strategic] systems. | ||
| + | - There will be strong incentives to build APS systems | (1). **80%** | ||
| + | - It will be much harder to develop APS systems that would be practically PS-aligned [power-seeking] if deployed, than to develop APS systems that would be practically PS-misaligned if deployed (even if relevant decision-makers don’t know this), but which are at least superficially attractive to deploy anyway | (1)–(2). | ||
| + | - Some deployed APS systems will be exposed to inputs where they seek power in misaligned and high-impact ways (say, collectively causing >$1 trillion 2021-dollars of damage) | (1)–(3). | ||
| + | - Some of this misaligned power-seeking will scale (in aggregate) to the point of permanently disempowering ~all of humanity | (1)–(4). | ||
| + | - This will constitute an existential catastrophe | (1)–(5). | ||
| + | |||
| + | The total AI risk is the product of the probabilities for each part of the model. | ||
| + | |||
| + | This same model has been used by Nate Soares and Eli Lifland to calculate their estimates of AI risk. Several other people in the table have also used similar models. | ||
| + | |||
| + | ==== Scenarios ==== | ||
| + | |||
| + | One example of describing possible scenarios to calculate the x-risk from AI is due to Peter Wildeford. He calculates x-risk from AI by 2100 by breaking it down in the following way: | ||
| + | |||
| + | - We get aligned and very capable AI systems, but we do not end the time of perils. We however still survive to the end of the century. | ||
| + | - Very capable AI systems are not developed this century. We survive until the end of the century. | ||
| + | - An unaligned rogue AI system creates a singleton. | ||
| + | - We get aligned and very capable AI systems that decisively end the time of perils. | ||
| + | - We get very capable AI systems that are broadly aligned with human values, but we create an existential risk anyways via abuse of nonhuman animals and/or digital minds. | ||
| + | - Some subset of humanity intentionally takes over the world via a very capable AI system. | ||
| + | - An unaligned rogue AI system causes complete human extinction. | ||
| + | - Some subset of humanity intentionally takes over the world via a very capable AI system and in the process ends up causing complete human extinction. | ||
| + | - Human extinction arises through a nuclear war that is started by accident. | ||
| + | - Human extinction arises through a nuclear war that is started intentionally. | ||
| + | - Some unknown unknown thing causes an existential risk. **0.05%** | ||
| + | - Nanotech causes an existential risk. **0.04%** | ||
| + | - An engineered pandemic causes an existential risk. **0.01%** | ||
| + | |||
| + | The total AI risk is the sum of (3), (5), (6), (7), & (8). | ||
| + | |||
| + | |||
| + | |||
| + | //Primary author: Jeffrey Heninger// | ||
arguments_for_ai_risk/quantitative_estimates_of_ai_risk.1686782242.txt.gz · Last modified: 2023/06/14 22:37 by jeffreyheninger
Page Tools
Except where otherwise noted, content on this wiki is licensed under the following license: CC0 1.0 Universal
