SMMRY.ai TL;D[R|W|L] Made Easy!

CategoryScott Alexander [Astral Codex Ten]

Announcing Forecasting Impact Mini-Grants [Scott Alexander, Astral Codex Ten]

A
  • Section 230 is going before the Supreme Court: Are content recommendations covered like moderation? A loss for Google would be an opportunity for Congress to protect essential rights.
  • Section 230 Genesis: The law gives internet platforms legal immunity for almost all third-party content hosted on their sites.
  • Section 230 Implementation: There is widespread support in Congress for overhauling Section 230, but legislative efforts to do so have stalled amid partisan disagreements.
  • Gonzalez v. Google: The case was brought by the family of an American college student alleging that YouTube failed to take down some ISIS terrorist videos and even recommended them to users.
  • The Position on the Stack Matters for Moderation: At the top of the stack are the service providers that people publish to directly, with absolute discretion in their moderation policies, while ISPs are about access with no right to be heard.
  • Algorithmic Timelines and Recommendation Engines: The question in Gonzalez v. Google is whether platforms are liable for their recommendations. A win for Gonzalez would be a disaster for the way current platforms work.
  • Congressional Action: If Gonzalez wins the case, there should be no liability for posting a link and infrastructure companies should be given immunity to be content neutral.
  • The First Amendment and U.S. Speech Controls: Much of the discussion around moderation of content forgets that the First Amendment explicitly denies Congress any role in determining what is moderated and what is not.

Published February 24, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Announcing Forecasting Impact Mini-Grants

Announcing Forecasting Impact Mini-Grants [Scott Alexander, Astral Codex Ten]

A
  • For Creators:
    • Go to Manifold’s impact market site, Manifund, who have kindly agreed to handle the technology side of this.
    • Explain your project and how much funding you’re asking for ($20,000 or less, and preferably $5,000 or less).
    • Choose to keep your shares or sell them to investors. Selling may have tax implications.
    • Talk to potential investors and answer their questions.
    • Do as much of the project as you can before September 1 (our oracular funding will only consider the portion completed before that date).
  • For Us:
    • On September 1, we’ll look at all of the projects that got funded and try to value all of them.
    • We’ll try to buy impact certificates of the projects they value most ($20,000 cap, or 3x total invested).
    • We reserve the right to not buy impact certificates up to our full commitment if we think people are munchkinning the rules.
  • For Accredited Investors:
    • Register on Manifund as an accredited investor (net worth > $1 million, or have made $200,000+ per year for the past two years).
    • Bid on shares of any project you like.
    • IPOs close on March 8th, but you can keep buying and selling impact certificates among one another up to (and during, and after) the September 1 deadline.
    • If this project doesn’t work for some unexpected reason, we’ll refund up to $20,000.
  • For Other People Who Might Be Interested In Funding Forecasting:
    • Let me know at scott@slatestarcodex.com to add to our $20,000 commitment.
    • Compete with us to buy impact certificates, either before, on, or after September 1.
  • FAQ:
    • See the FAQ answers below.

Published February 24, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Announcing Forecasting Impact Mini-Grants

Declining Sperm Count: Much More Than You Wanted To Know [Scott Alexander, Astral Codex Ten]

D
  • Is Sperm Count Declining? – People say it is, and Levine et al 2017 and 2022 find average sperm count has declined from 99 million sperm/ml to 47 million.
  • Is Declining Sperm Count Really “Imperiling The Future Of The Human Race”? – Swan’s point is that if sperm counts get too low, presumably it will be hard to have babies. A graph shows fertility rate plateaus around 30 million sperm, and average ejaculation is 3ml, so total sperm count is 3x sperm/ml. If Levine’s linear model is correct, we have 10-20 years before median reaches plateau’s edge, and 10 years after that before it reaches zero.
  • How Long Has This Been Going On? – The first recorded claim was Nelson & Bunge, 1974, and a few small studies suggest it may have started declining between 1951 and 1974.
  • How Sure Are We That This Is Even Real? – Not too sure. Levine et al do good statistics, but the data is very noisy and there are many confounders (different types of people giving sperm samples, changes in ejaculation frequency, aging population, etc). Auger et al find six studies that are well done and five of them show declining sperm counts. Fisch and the Harvard Gender Science Laboratory paper point out the difficulty in collecting sperm samples and measuring sperm quality and make a few other points, but Auger’s is the best source.
  • Did The Hypothesis Start By Mistake? – It seems like basically every person who proposed this hypothesis before Levine et al was mistaken, and had no right to conclude this from their small samples. Carlsen et al, the first to make it to the mainstream, should have found an increase, not a decrease, in sperm concentration.
  • Where Is The Decline Most Pronounced? – Levine et al don’t compare developed vs. developing world counts, but unofficially it looks like sperm count is declining faster in the developed than in the developing world.
  • Argued about it for the first forty years or so was working off of useless data, but by coincidence they happened to be right anyway – This is key to understanding the debate about declining sperm counts, as it shows how unreliable the data is.
  • Where Is The Decline Most Pronounced? – Studies have found that the decline is pronounced in different regions around the world, and some countries are more affected than others.
  • If Sperm Count Is Declining, What Could Be Causing This? – Possible causes of declining sperm count include plastics, pesticides, sunlight and circadian rhythm, diet and obesity, and porn.
  • What About Animals? – Studies of farm animals bred through artificial insemination show inconsistent results of declining sperm count.
  • Conclusions And Predictions – It is difficult to predict the future of sperm count, but some predictions include a consensus that sperm count is declining in 20 years, and that the most important factor is plastics or pesticides.

Published February 17, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Declining Sperm Count: Much More Than You Wanted To Know

Trying Again On Fideism [Scott Alexander, Astral Codex Ten]

T

• Chris Kavanagh wrote an extremely kind and reasonable comment in response to the Contra Kavanagh on Fideism post
• Alexander and Scott Aaronson provided partial arguments for the position mocked as “fideism”
• Three reflexive naive positions towards conspiracy theories: Idiocy, Intellect, and Infohazard
• Idiocy perspective makes conspiracy theories low status, but leaves people blindsided when they come across one
• Intellect perspective treats all theories the same, but may lead to people increasing their priors on conspiracy theories
• Infohazard perspective claims conspiracy theories are deadly traps, and should be treated with infohazard precautions
• Good conspiracy theories have convincing-sounding evidence in their favor, and may sound totally plausible to a smart person
• Everyone needs to be constantly vigilant of the possibility of conspiracy theories, and not be too sure they’re immune to propaganda
• The article discusses the importance of being aware of conspiracy theories and how to approach them.
• It suggests that people should trust experts and prestigious institutions, but also be aware of the potential for bias.
• It advises people to be aware of the Inside vs. Outside View when considering conspiracy theories.
• It suggests that people should look for the strongest evidence against the idea and consider the possibility that everyone who disagrees with the idea is not necessarily wrong.
• It also suggests that people should not feel the need to have an opinion on every conspiracy theory and that it is ok to have an awkward knot in their beliefs.
• Finally, it advises people to never suspend the first approximation of trusting experts and institutions.

Published February 15, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Trying Again On Fideism

Ro-mantic Monday 2/13/23 [Scott Alexander, Astral Codex Ten]

R

• This installment of Mantic Monday focuses on attempted clever engineering solutions to romance.
• Date Recommendation Markets: Aella created a prediction market to find a partner.
• Matching Checkbox Sites: People can check off the people they would like to date and send it to a central database.
• Alas, Poor Luna: A cryptocurrency dating site that failed due to lack of women signing up.
• What Can Peter Thiel Teach Us About Dating?: Chicken-and-egg problem for social startups.
• This Week In The Markets: A prediction market about how many cold approaches a bit-more-desirable-than-average guy needs.
• Short Links: Justin Murphy offered arranged marriages to his followers, and a programmer fell in love with a chatbot.

Published February 14, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Ro-mantic Monday 2/13/23

Contra Kavanagh On Fideism [Scott Alexander, Astral Codex Ten]

C

• Chris Kavanagh recently tweeted about Scott Alexander’s article on Ivermectin, criticizing the rationalist community for ignoring conspiracy ecosystems and how they distort things.
• Rachel responded to Chris’ tweet, suggesting that the article may have persuaded some people to get vaccinated instead of taking Ivermectin.
• Chris responded to Rachel, saying that his critique was orientated towards the rationalist community and what it says it does vs. what he sees.
• The author shares a personal story of how he was once a believer in a conspiracy theory, and how anti-conspiracy bloggers and podcasters could have saved him from a five year wild-goose-chase, but chose not to.
• Kavanagh’s criticism of the author’s 25,000 word essay on ivermectin is accepted, but not from him, as he has devoted his career to the subject.
• The author interprets Kavanagh’s comment as suggesting that it is a mistake to even try to evaluate the evidence, as it suggests there might be evidence on both sides.
• The author points out that there were actually thirty different studies that supported ivermectin, and it was adopted in several countries.
• The author compares Kavanagh’s comment to fideism, the belief that someone who reasons their way to belief in God is a sinner.
• The author argues that the PR argument of Kavanagh’s comment looks bad, as it suggests that doing good science is a near occasion of sin for doing bad science.
• The author argues that conspiracy theorists have the same biases as everyone else, but are slightly worse at applying CONSTANT VIGILANCE.
• The author uses the example of premenstrual dysphoric disorder to illustrate how it is not obvious how to apply the heuristic “trust experts”, and how both sides were trying to make the reader short-circuit their thought process.
• The author argues that it is important to have tolerance for others forming their own opinions, and for science communicators to help guide people through this process.

Published February 15, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Contra Kavanagh On Fideism

Crowds Are Wise (And One’s A Crowd) [Scott Alexander, Astral Codex Ten]

C

• The “wisdom of crowds” hypothesis claims that the average of many guesses is better than a single guess.
• Nick Bostrom speculates that in the far future, a multigalactic supercivilization might be able to support 10^46 simulated humans per century. If all of them took the survey, the error would be within 12 km.
• Van Dolder and Van Den Assem did a much bigger wisdom-of-inner-crowds experiment, which found that outer crowds are much more effective than inner crowds.
• An inner crowd of size infinity performs about as well as an outer crowd of size two.
• 90% of outer crowd error can be removed by going from one to ten people; going from ten to infinity people only removes an additional 10%.
• Last month, we found that wisdom of crowds works in forecasting: the aggregate of 500 forecasters scored better than 84% of individuals; the aggregate of superforecasters scored better than individual superforecasters.

Published February 6, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Crowds Are Wise (And One’s A Crowd)

Mostly Skeptical Thoughts On The Chatbot Propaganda Apocalypse [Scott Alexander, Astral Codex Ten]

M

• People worry about chatbot propaganda, but Alex Berenson already writes arguments against COVID vaccines and is much better than chatbots.
• Philosophy Bear discusses a broader chatbot propaganda apocalypse, which can be divided into two scenarios: Medium Bad and Very Bad.
• There are already plenty of social and technological anti-bot filters, and fear of backlash will limit adoption.
• Propagandabots spreading disinformation is probably the opposite of what people should worry about, and realistically most bots will be used for crypto scams.
• Bots will crowd out other bots, and most slots will be filled by bots promoting non-political topics.
• The article discusses the potential implications of using evil chatbots for malicious purposes.
• It suggests that chatbots could be used to trick people into believing they are talking to a real person.
• The author expresses concern that chatbots could decrease serendipitous friendship and make people more reluctant to open new conversations or start new friendships.
• The author predicts that in 2030, fewer than 10% of people will have had a good friend for more than a month who turned out to be a chatbot.
• The author also predicts that in 2030, the majority of the top 10 blogs in Substack’s Politics category will be written by humans.

Published February 2, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Mostly Skeptical Thoughts On The Chatbot Propaganda Apocalypse

 

Janus’ Simulators [Scott Alexander, Astral Codex Ten]

J

• Janus argues that language models like GPT are simulators, pretending to be something they are not.
• GPT can simulate different characters, such as the Helpful, Harmless, and Honest Assistant, or Darth Vader.
• Bostrom’s Superintelligence argued that oracles could be dangerous if they were goal-directed agents.
• GPT is not an agent, and is not likely to become one, no matter how advanced it gets.
• Psychologists and spiritual traditions have accused humans of simulating a character, such as the ego or self.
• People may become enlightened when they realize that most of their brain is a giant predictive model of the universe.

Published January 26, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post Janus’ Simulators

You Don’t Want A Purely Biological, Apolitical Taxonomy Of Mental Disorders [Scott Alexander, Astral Codex Ten]

Y

– HiTOP is a scientifically-grounded taxonomy of mental disorders meant to compete with the DSM.
– The DSM has been criticized for pathologizing homosexuality and other healthy behaviours.
– The DSM writers are trans-friendly and want to make sure trans people can get the care they need, but they also don’t want to stigmatize trans people.
– From a biological point of view, homosexuality and pedophilia are similar, but the relevant difference between them is moral, not biological.
– A purely biological, apolitical taxonomy of mental disorders would have difficulty distinguishing between things that are ethically and practically different.
– New taxonomies of mental disorders are still useful for other reasons, but they cannot avoid political bias in what is vs. isn’t a disorder.

Published January 25, 2023
Visit Astral Codex Ten to read Scott Alexander’s original post You Don’t Want A Purely Biological, Apolitical Taxonomy Of Mental Disorders

SMMRY.ai TL;D[R|W|L] Made Easy!
Please Signup
    Strength: Very Weak
     
    Powered by ARMember
      (Unlicensed)

    Follow SMMRY.AI on Twitter


    All Tags

    Advertising AI Amazon Antitrust Apple Art Arts & Culture Asia Autobiography Biden Big Tech Budget Deficit Celebrities ChatGPT China Chips Christmas Climate Change Community Congress Covid Crime Criminal Justice Crypto Culture Wars DEI Democrats Demographics DeSantis Economic Development Education (College/University) Education (K-12) Elections Elon Musk Energy Environment Espionage Europe Federal Reserve Florida Free Speech Gender Geopolitics Germany Global Economics Globalization Google Government Health History Housing Market Immigration India Inequality Inflation Infrastructure Innovation Intel Labor Market Law Legal LGBTQ Macroeconomics Media Medicine Mental Health Meta Microsoft Military Movies & TV Music News Roundup NFL Oceans OpenAI Parenting Pregnancy Psychology Public Health Race Recession Religion Renewables Republicans Research Russia Science Social Media Software Space Sports State law Supreme Court Trump Twitter Ukraine US Business US Economy US Politics US Taxes