Scott Alexander gave a 90% probability that he would remember and publicly grade these predictions sometime in 2023.
Resolved true The record shows it happened or was so.
Prediction: Statements about events outside the person's control, checked once they fell due: did they come true?
What would count as true or false
This is a probabilistic forecast; judge whether the event happened. True if, at some point in calendar 2023, Scott Alexander publicly graded the predictions from his February 2018 post (the five-year predictions), e.g. in a blog post scoring them. Mostly true if the grading was published in 2023 but covered most rather than all of the predictions. Mostly false if he graded them, but only after 31 December 2023, or only in passing without real grading. False if a 2023 or later record shows no grading in 2023, e.g. he graded them in 2024 or said he had forgotten.
Set before the research began.
Evidence (4)
-
E1 Grading My 2018 Predictions For 2023 Astral Codex Ten (Scott Alexander) · 20 February 2023 Own record
“Last week was the tenth anniversary of my old blog (I accept your congratulations), so it's time to look back on my terrible doomed 2018 predictions and see how I did at predicting the last half decade, starting with:”
-
E2 Grading My 2018 Predictions For 2023 Astral Codex Ten (Scott Alexander) · 20 February 2023 Own record
“I grade 1, 5, and 6 as coming true; 2, 3, and 4 as not coming true, and 7 as “more”, all of which I got directionally correct.”
-
E3 Grading My 2018 Predictions For 2023 Astral Codex Ten (Scott Alexander) · 20 February 2023 Own record
“My biggest failure here was 10, where I gave Roe vs. Wade only a 1% chance (!) of being overturned!”
-
E4 Grading My 2018 Predictions For 2023 — Scott Alexander, curated ACX Index (unofficial fan index, not affiliated with Scott Alexander) · 20 February 2023
“Scott grades his 2018 five-year predictions across AI, world affairs, US culture/politics, economics, science/tech, and x-risks (best call: AI generating images/stories to a prompt; worst: giving Roe's overturn only 1%), then lays out fresh 2028 predictions”
Reasoning
E1 is a post titled "Grading My 2018 Predictions For 2023", dated 2023-02-20, in which Scott Alexander says it is time to look back on his 2018 predictions. E2 and E3 show him scoring specific numbered predictions, including the Roe v. Wade forecast. E4 is an independent fan index, and it describes the same post as grading his 2018 five-year predictions across AI, world affairs, US culture and politics, economics, science and tech, and x-risks. So the event he gave 90% odds to did happen in calendar 2023.
Cite this claim
Scott Alexander, “Scott Alexander gave a 90% probability that he would remember and publicly grade these predictions sometime in 2023.” (said 15 February 2018). Verdict: resolved true. Reliability API, claim c_4d7337c4c4b1, method 0.1.0+8bea457b, dataset release v1.1.2. https://epistacklabs.com/claims/c_4d7337c4c4b1
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
More prediction claims by Scott Alexander
-
Mostly true
“Our scenario focuses on coding in particular because we think coding is what starts the intelligence explosion.”
The authors of the scenario think coding is what will start the intelligence explosion.
-
Mostly true
“We expect that as the AI labs become more capable, they tell the government about this because they want government contracts, they want government support.”
-
Mostly true
“In our scenario, that starts with cyber warfare, the government sees that these AIs are now as capable as the best human hackers, but can be deployed at humongous scale.”
-
Resolved true
“So they become extremely interested and they discuss nationalizing the AI companies.”
-
Mostly true
“I expect scaling to continue to produce gains, but this is less obvious than it’s been the past five years.”
Scaling AI models will continue to produce gains, though less obviously than in the past five years.
Verdicts are made automatically; how verdicts are made.