Lunch at the Labs

Various web links I found to be of interest recently.

• DeepFakes, Can You Spot Them?

  At Detect Political Fakes, we will show you a variety of media snippets (transcripts, audio, and videos). Half of the media snippets are real statements made by Joseph Biden and Donald Trump. The other half of the media snippets are fabricated. The media snippets that are fabricated are produced using deepfake technology. We are asking you to share how confident you are that a media snippet is real or fabricated.

• Pandora Papers: Stop the enablers that help billionaires dodge taxes

  The Pandora Papers disclosures reveal the elaborate mechanisms that the wealthy deploy to shift funds between global jurisdictions, masking their true wealth and minimizing their tax obligations. It unmasks the U.S. as a tax haven - including the state of South Dakota with its proliferation of dynasty trusts.
  ...
  While we should hold the billionaire tax dodgers to account, not enough scrutiny is focused on the enablers, what social scientists describe as "the wealth defense industry." These are the tax attorneys, accountants, wealth managers and family-office staffers that are paid millions to help billionaires sequester trillions.
  ...
  The wealth defenders will crow "everything we do is legal" but segments of the wealth defense industry are implicated in writing the rules governing trusts in many U.S. states and designing the exotic transactions that dance along the intent of the law. While the left hand is using various wealth-hiding techniques and loopholes, the right hand is fending off oversight, lobbying for special treatment, and creating new loophole innovations.

• The Advantage of Permission & The Fall Of Oligarchies

  Ian Welsh

  Regulatory hurdles are massive. Everyone knows, for example, how to make insulin, and you'd think someone would get into the business and undercut producers. Insulin costs about $4 to make, and sells for over $300. In economics this looks like the sort of situation that would automatically lead to competition.

  It doesn't. For one, getting permission to produce is hard and expensive. For the second, if you did manage to, it would be costly to setup the initial factory and supply chain, and that means the current producers would simply step in and undercut the new entrant till they went out of business (people would buy the cheapest), then push prices back up once they had driven the challenger out of business.

• Rethinking General Anesthesia

  Emery N. Brown, M.D., Ph.D.

  Consequently, they have developed a principled, neuroscience-based paradigm for using the EEG to monitor the brain states of patients receiving general anesthesia. In addition, they demonstrate that the state of general anesthesia can be rapidly reversed by activating specific brain circuits. Finally, they show that the state of general anesthesia can be controlled using closed-loop feedback control systems.

  Some things I remember from the talk:

  • The state it puts your brain is like a coma and not being asleep.
  • Anesthesiologists should use EEG's to monitor brain waves but all they do is monitor things like breathing, pulse, etc He said only about 25% of them do and it's been difficult to get them to do something they haven't been doing their whole life. (IMO it's like how long it took doctors to wash their hands and wear gloves).
  • Instead of just letting patients wait until the anesthesia wears off, things can be done to cause waking up. In general people are kept anesthetized too long.
  • Multimodal General Anesthesia uses more than one drug but less of each, increasing the likelihood of desired effects and reducing the likelihood of bad side effects.

• Risk of severe breakthrough Covid-19 higher for seniors and people with underlying conditions

  As of August 30, the US Centers for Disease Control and Prevention has received reports of 12,908 severe breakthrough cases of Covid-19 among fully vaccinated people that resulted in hospitalization or death. For the more than 173 million people who were fully vaccinated by that date, that represents a less than a 1 in 13,000 chance of experiencing a severe breakthrough case of Covid-19.

  About 70% of breakthrough cases resulting in hospitalization were among adults 65 and older and about 87% of breakthrough cases resulting in death were among adults 65 and older, the CDC data suggests.

  Also see How much less likely are you to spread covid-19 if you're vaccinated?.

  A recent study found that vaccinated people infected with the delta variant are 63 per cent less likely to infect people who are unvaccinated.
  What is important to realise, de Gier says, is that the full effect of vaccines on reducing transmission is even higher than 63 per cent, because most vaccinated people don’t become infected in the first place.

• New evidence undermines the COVID lab-leak theory - but the press keeps pushing it

  Despite mounting evidence that the virus reached humans through natural pathways - from infected animals such as bats - the lab-leak hypothesis recently jumped back into the news, thanks to CNN, the investigative news site the Intercept, and the Atlantic.

  All treat the idea that the virus escaped from a lab credulously. They downplay or entirely ignore the latest scientific findings that support the theory that the virus' origin can be found in the animal kingdom - the view accepted by a preponderance of experts in virology.

  Also see Why I Still Believe Covid-19 Could Not Have Originated in a Lab.
  Lab escape theories cannot clearly account for a virus that has evolved for human-to-human transmissibility.

• The Triumph and Terror of Wang Huning

  A member of the CCP's seven-man Politburo Standing Committee, he is China's top ideological theorist, quietly credited as being the "ideas man" behind each of Xi's signature political concepts, including the "China Dream," the anti-corruption campaign, the Belt and Road Initiative, a more assertive foreign policy, and even "Xi Jinping Thought." Scrutinize any photograph of Xi on an important trip or at a key meeting and one is likely to spot Wang there in the background, never far from the leader's side.
  ...
  "Socialism with Chinese Characteristics" has rapidly transformed China into one of the most economically unequal societies on earth. It now boasts a Gini Coefficient of, officially, around 0.47, worse than the U.S.'s 0.41. The wealthiest 1% of the population now holds around 31% of the country's wealth (not far behind the 35% in the U.S.). But most people in China remain relatively poor: some 600 million still subsist on a monthly income of less than 1,000 yuan ($155) a month.

  Meanwhile, Chinese tech giants have established monopoly positions even more robust than their U.S. counterparts, often with market shares nearing 90%. Corporate employment frequently features an exhausting "996" (9am to 9pm, 6 days a week) schedule. Others labor among struggling legions trapped by up-front debts in the vast system of modern-day indentured servitude that is the Chinese "gig economy." Up to 400 million Chinese are forecast to enjoy the liberation of such "self-employment" by 2036, according to Alibaba.

• Futurists have their heads in the clouds

  Erik Hoel, On making good predictions for 2050

  Why are such smart people so impossibly bad at this? Why do they jump to your "headwires" controlling your "robot cook" as you get ready to drive your floating car? The truth is that most futurists are attracted to speculating about the future for the same reason as science fiction writers: they like geeking out about technological possibilities and the associated metaphysics. That's how you get a sci-fi book without the plot.

  If you want to predict the future accurately, you should be an incrementalist and accept that human nature doesn't change along most axes. Meaning that the future will look a lot like the past. If Cicero were transported from ancient Rome to our time he would easily understand most things about our society.

• Whole Brain Emulation: No Progress on C. elgans After 10 Years
  1. Knowing the connections isn't enough, we also need to know the weights and thresholds. We don't know how to read them from a living worm.
  2. C. elegans is able to learn by changing the weights. We don't know how weights and thresholds are changed in a living worm.

  The best we can do is modeling a generic worm - pretraining and running the neural network with fixed weights. Thus, no worm is "uploaded" because we can't read the weights, and these simulations are far from realistic because they are not capable of learning. Hence, it's merely a boring artificial neural network, not a brain emulation.

• Decreasing Number of Oil Spills

  Over the past four and half decades - the time for which we have data - oil spills from tankers decreased very substantially. The dataset by the International Tanker Owners Pollution Federation (ITOPF) covers more than four decades over which time the incidence of large oil spills from tankers greatly decreased. While in the 1970s there were 24.5 large (> 700 tonnes) oil spills per year, in the 2010s the average number of large oil spills decreased to 1.7 oil spills per year.

• Grocery Store Opens 'Chat Registers' for Lonely Customers

  A Dutch supermarket chain says it will introduce 200 "chat registers" in its stores for customers who aren't in a hurry and want to have a chat during checkout.
  ...
  "The Kletskassa is a checkout especially for people who are not in a hurry and feel like having a chat," a Jumbo representative wrote in an email to Motherboard. "Many people, especially the elderly, sometimes feel lonely. It's a small gesture, but a very valuable one, especially in a world that is digitizing and getting faster and faster."

Some recent links related to Covid-19.

• The tangled history of mRNA vaccines

  Hundreds of scientists had worked on mRNA vaccines for decades before the coronavirus pandemic brought a breakthrough.

• Hundreds of AI tools have been built to catch covid. None of them helped.

  In June, the Turing Institute, the UK's national center for data science and AI, put out a report summing up discussions at a series of workshops it held in late 2020. The clear consensus was that AI tools had made little, if any, impact in the fight against covid.

• New evidence undermines the COVID lab-leak theory - but the press keeps pushing it

  "That a laboratory leak would find its way to the very place where you would expect to find a zoonotic transmission is quite unlikely," Joel Wertheim, an associate professor at UC San Diego's medical school, told me. "To have it find its way to multiple markets, the exact place where you would expect to see the introduction, is unbelievably unlikely."
  ...
  Evidence that artificial manipulation of a virus gave rise to SARS-CoV-2 has faded, as scientists find more evidence that features of SARS-CoV-2 thought to be unnatural occur in nature. Meanwhile, evidence for zoonotic transmission is constantly accumulating. No one who reports on the issue without acknowledging these two trends should be trusted.

• Experts: COVID-19 infection offers less protection than many think

  Your case of COVID-19 -- especially if it was very mild -- probably didn't create enough of an immune response to provide lasting protection against the coronavirus, said Dr. Buddy Creech, president-elect of the Pediatric Infectious Diseases Society.

  Even if you've had COVID-19, you need to get vaccinated to make sure yous don't catch a second case that might be worse than your first, experts said.

• Virus fears linger for vaccinated older adults

  In a sign of the starkly different way Americans view the coronavirus pandemic, vaccinated older adults are far more worried about the virus than the unvaccinated and far likelier to take precautions despite the protection afforded by their shots, according to a new poll out Wednesday from The Associated Press-NORC Center for Public Affairs Research.
  ...
  That worry is taking a toll: Those concerned about COVID-19 are less likely to rate their quality of life, mental and emotional health, and social activities and relationships as excellent or very good.

• Is the Delta variant really more than twice as transmissible as the original strain of the virus?

  The goal of this essay is not to convince you that Delta isn't more transmissible than the previously established strains. To be clear, I think it probably is (SARS-CoV-2 is subject to natural selection and it's not particularly surprising that more transmissible variants should emerge over time), but I also think its transmissibility advantage has almost certainly been vastly overestimated. The view that Delta is more than twice as transmissible as the original strain is superficially plausible because it comes with a compelling narrative.
  ...
  The evidence that has been adduced for this specific claim is no more convincing. As we have seen, the estimates of Delta's transmissibility advantage that everybody uncritically cites could easily be misleading, and there are good reasons to think that Delta's transmissibility advantage is actually much lower.

• The Myth of 'Long COVID'

  Disclaimer: This is from a far-right / conservative author/publication and I haven't verified the claims, so read with caution.

  The largest study so far of "long-haulers," published by researchers at University College London in July, comprised nearly 4,000 subjects from over 56 countries. Participants were over the age of 18 and suffered from symptoms lasting at least 28 days. The researchers acknowledged merely in passing that in the study a mere 27% or 1020 of these "COVID long-haulers" had evidence of exposure to the SARS-CoV-2 virus. That's whether antigen or antibody. The only connection to COVID was the attestation of the sufferers. They "felt" they had COVID, regardless of evidence.

  An August study of 3,151 British ("long haulers") in Pragmatic and Observational Research found only 17.2% were test-confirmed positive. A further 12% said they were told they had acute COVID, but no test was performed. And over 70% admitted it was merely self-diagnosis. An influential and scary article in the Atlantic reported some two-thirds of "long-hauler" patients had negative coronavirus antibody tests without making the obvious inference. An advocacy group study released in May 2020 found that only "About a quarter of respondents (23.1%) tested positive for COVID-19" but in "our analysis, we included all responses regardless of testing status."

  For some contrary evidence see,
  At least one long-term symptom seen in 37% of COVID-19 patients -study.
  

• No, Vaccinated People Are Not 'Just as Likely' to Spread the Coronavirus as Unvaccinated People

  So let me make one thing clear: Vaccinated people are not as likely to spread the coronavirus as the unvaccinated. Even in the United States, where more than half of the population is fully vaccinated, the unvaccinated are responsible for the overwhelming majority of transmission.
  ...
  In the aftermath of the Provincetown announcement, many who had gotten their shots were confused about what the news meant for them, especially when headlines seemed to imply that vaccinated individuals are as likely to contract and transmit COVID-19 as the unvaccinated. But this framing missed the single most important factor in spreading the coronavirus: To spread the coronavirus, you have to have the coronavirus. And vaccinated people are far less likely to have the coronavirus-period. If this was mentioned at all, it was treated as an afterthought.

• Stanford students are more likely to wear masks on bicycles than helmets

  I think this anecdote is instructive in understanding the social dynamics that have emerged in the COVID-19 pandemic. Seemingly intelligent and well-rounded people (Stanford students, for example) have adopted bizarre, pointless habits to comport with new expectations about how to "stay safe" -- like wearing masks outdoors -- all while continuing in much more risky behaviors. This is not to say that riding a bicycle without a helmet is *especially* risky, or that I believe helmets should be mandated (they shouldn't). But it's absolutely a bigger risk than COVID-19 for a vaccinated twenty-something.