How to avoid getting lost reading Scott Alexander and his 1500+ blog posts? This unaffiliated fan website lets you sort and search through the whole codex. Enjoy!

See also Top Posts and All Tags.

Minutes:
Blog:
Year:
4542 tags
Show all filters
4 posts found
Sep 12, 2022
acx

I Won My Three Year AI Progress Bet In Three Months

12 min • 1,502 words • 313 comments • 159 likes • podcast
Scott Alexander won his three-year bet on AI progress in image generation compositionality within three months, demonstrating unexpectedly rapid AI advancement. Longer summary
Scott Alexander discusses his recent bet about AI image models' ability to handle compositionality, which he won much sooner than expected. He explains the concept of compositionality in AI image generation, showing examples of DALL-E2's limitations. Scott then describes the bet he made, predicting improvement by 2025. However, newer AI models like Google's Imagen achieved the required level of compositionality within just three months. This rapid progress supports the idea that AI development is advancing faster than many predict, and that scaling and normal progress can solve even complex problems in AI. Shorter summary
Mar 28, 2022
acx

Information Markets, Decision Markets, Attention Markets, Action Markets

24 min • 3,044 words • 118 comments • 76 likes • podcast
Scott Alexander explores different types of prediction markets, their uses, limitations, and ethical considerations in decision-making and information gathering. Longer summary
Scott Alexander discusses various types of prediction markets, including information markets, decision markets, attention markets, and action markets. He explores their potential uses, limitations, and ethical considerations. The post covers how these markets can be used to predict past and present events, guide decision-making, allocate attention to important issues, and incentivize actions. Scott also discusses the challenges and potential pitfalls of each type of market, such as trust issues, the need for resolution, and potential for abuse. He concludes by drawing parallels between prediction markets and concepts in AI safety and neuroscience. Shorter summary
Jan 17, 2017
ssc

Another Followup To “Economists On Education”

6 min • 702 words • 203 comments • podcast
Scott Alexander's bet about readers misinterpreting an article on economists' views on school vouchers is settled through two surveys, confirming his argument. Longer summary
Scott Alexander follows up on his previous post about economists' views on school vouchers as reported in a news article. He had argued that the article misrepresented economists' opinions, and to settle the disagreement, he proposed a bet that most people would misinterpret the article. Two independent surveys were conducted on Mechanical Turk, both showing that a large majority of readers (91% and 78%) misinterpreted the article to claim that economists opposed privatizing education. This empirically settled the disagreement, with Scott winning the bet. He expresses satisfaction at being able to decisively resolve a dispute through experimentation. Shorter summary
Dec 31, 2016
ssc

Addendum To “Economists On Education”

8 min • 1,023 words • 86 comments • podcast
Scott Alexander argues that a New York Times article misrepresented economists' views on school vouchers, offering evidence and proposing a bet to prove the article was misleading. Longer summary
Scott Alexander critiques a New York Times article that claimed economists generally don't support school vouchers. He argues the article misrepresented survey data and gave a misleading impression. Scott points out that the data could be interpreted to support the opposite conclusion, that the article's phrasing likely led readers to an incorrect understanding of economists' views, and that a follow-up survey showed much stronger economist support for vouchers. He offers to bet that most readers would get a false impression from the article, initially agreeing to a bet with Noah Smith, though this later fell through. Shorter summary