Back
Epoch AI: Literature Review of TAI Timelines
webCredibility Rating
4/5
High(4)High quality. Established institution or organization with editorial oversight and accountability.
Rating inherited from publication venue: Epoch AI
A useful meta-level survey for those wanting a structured comparison of TAI/AGI timeline forecasts; helpful for situating individual models like Biological Anchors or semi-informative priors within the broader forecasting landscape as of early 2023.
Metadata
Importance: 72/100blog postanalysis
Summary
Epoch AI reviews and compares quantitative models and expert judgment-based forecasts predicting when transformative AI will arrive, including biological anchors, semi-informative priors, and prediction market aggregates. Inside-view models tend to predict shorter timelines (median ~2052) while outside-view models predict longer timelines (median >2100), with expert judgment forecasts often more aggressive than either. The review also provides Epoch AI team's subjective weightings on the relative trustworthiness of each approach.
Key Points
- •Inside-view models like Cotra's Biological Anchors give a median TAI arrival around 2052; outside-view models like Davidson's semi-informative priors give medians beyond 2100.
- •Judgment-based forecasts (e.g., Samotsvety's median of 2043) tend to align more with inside-view models and are often more aggressive.
- •Epoch AI rates Cotra's Biological Anchors as most compelling inside-view model, Davidson's semi-informative priors as best outside-view, and Samotsvety as best judgment-based forecast.
- •The review distinguishes 'prior-forming' model-based forecasts from 'posterior-forming' judgment-based forecasts, aggregating internal team credences for each.
- •Different sources operationalize AGI/TAI differently, limiting direct comparisons across forecasts.
Cited by 1 page
| Page | Type | Quality |
|---|---|---|
| Long-Timelines Technical Worldview | Concept | 91.0 |
Cached Content Preview
HTTP 200Fetched Apr 7, 202628 KB
Literature review of transformative artificial intelligence timelines | Epoch AI
Previous work: Grokking “Forecasting TAI with biological anchors“ , Grokking “Semi-informative priors over AI timelines”
Highlights
The review includes quantitative models, including both outside and inside view, and judgment-based forecasts by (teams of) experts.
While we do not necessarily endorse their conclusions, the inside-view model the Epoch AI team found most compelling is Ajeya Cotra’s “Forecasting TAI with biological anchors” , the best-rated outside-view model was Tom Davidson’s “Semi-informative priors over AI timelines” , and the best-rated judgment-based forecast was Samotsvety’s AGI Timelines Forecast .
The inside-view models we reviewed predicted shorter timelines (e.g. bioanchors has a median of 2052) while the outside-view models predicted longer timelines (e.g. semi-informative priors has a median over 2100). The judgment-based forecasts are skewed towards agreement with the inside-view models, and are often more aggressive (e.g. Samotsvety assigned a median of 2043).
Introduction
Over the last few years, we have seen many attempts to quantitatively forecast the arrival of transformative and/or general Artificial Intelligence (TAI/AGI) using very different methodologies and assumptions. Keeping track of and assessing these models’ relative strengths can be daunting for a reader not familiar with the field. As such, the purpose of this review is to:
Provide a relatively comprehensive source of influential timeline estimates, as well as brief overviews of the methodologies of various models, so readers can make an informed decision over which seem most compelling to them.
Provide a concise summarization of each model/forecast distribution over arrival dates.
Provide an aggregation of internal Epoch AI subjective weights over these models/forecasts. These weightings do not necessarily reflect team members’ “all-things-considered” timelines, rather they are aimed at providing a sense of our views on the relative trustworthiness of the models.
For aggregating internal weights, we split the timelines into “model-based” and “judgment-based” timelines. Model-based timelines are given by the output of an explicit model. In contrast, judgment-based timelines are either aggregates of group predictions on, e.g., prediction markets, or the timelines of some notable individuals. We decompose in this way as these two categories roughly correspond to “prior-forming” and “posterior-forming” predictions respectively.
In both cases, we elicit subjective probabilities from each Epoch AI team member reflective of:
how likely they believe a model’s assumptions and methodology to be essentially accurate, and
how likely it is that a given forecaster/aggregate of forecasters is well-calibrated on this problem,
respectively. Weights are normalized and linearly aggregated across the team to arrive at a summary probabi
... (truncated, 28 KB total)Resource ID:
2cb4447b6a55df95 | Stable ID: sid_9YeS2VeW1p