Dr Jeyanthi Kunadhasan is an Australian Anaesthetist that has been on this COVID mRNA vaccine is bad rant for a long time, and has been widely discredited in Australia. She is misinterpreting the deaths data in the trial, which unfortunately is a widespread naive error.
1. In the part of the trial in which you can actually compare raw death counts because of the randomization, the deaths were 15-14, with confidence intervals nearly completely overlapping, and so there was no evidence vaccine arm death rates were significantly higher than placebo arm.
2. After the subjects were unblinded, the placebo subjects were allowed to crossover to the vaccine arm, and the vast majority did, with >>90% of the subjects vaccinated in the end and <<10% remaining vaccinated. After crossover, there were 9 additional deaths, 6 of who had been vaccinated and 3 of which had not yet been vaccinated.
3. But one cannot simply add the 6 to the 15 and 3 to the 14 and compare 21 vs. 17 to compare death rate of vaccinated to unvaccinated, since after the crossover, the vast majority were vaccinated and very few remained unvaccinated.
4. If one wants to assess the death rate after vaccination during the post-crossover period, they must treat it as an observational cohort study with time-varying vaccination status, adjusting for the wildly different group sizes and the fact that there were far more person-years of potential risk of death (i.e. denomiantors for death rates) after vaccination than in the pre-vaccination group. If such an analysis is done, there is also no evidence that death rate after vaccination is greater than before vaccination.
So, in the end, there is no evidence that the death rate in the vaccinated group is greater than the placebo/unvaccinated group.
Then her second big error is presuming, without any evidence whatsoever, that the excess deaths in Australia are predominately caused by the vaccine.
Googles weather prediction engine is already very good, and the new hurricane model was breathtakingly good this season when tested against actual hurricane paths. Meanwhile, the US Government Global Forecasting System continues to get worse.
But to expand: the US flagship forecast model just had its worst year predicting hurricanes since 2005. The trend of errors over the last few years hasn't been great.
I disagree with your logic. Increased mean error 72 hours out (vs 24 hours out) is not an indication that GFS is getting worse over time. At that scale it’s obviously getting better over time; 24 hours out is further in the future than 72 hours out.
However, an increase in the mean error at the same time out year over year (or between 2005 and 2025) is an indication of an issue, and that’s what we see.
Fi Fi Lee also recently founded a new AI startup called World Labs, which focus on creating AI world models with spatial intelligence to understand and interact with the 3D world, unlike current LLM AI that primarily processes 2D images and text. Almost exactly the same focus as Yann LeCun's new venture stated in the parent article.
Americans are very weird when it comes to metric. They often quote mobile phones as having something like "a six inch screen size but now only 12mm thick" - pick a measurement system people !
After switching to the metric system ('70-80s) some things are still measures in imperial units. If you slice some ham at a counter in a grocery store, it's in grams. You then turn around and get a pound of apples and a gallon of milk. Nuts are in grams, and soda is in liters. Also the body weight tends to be in pounds. Tools are both metric and imperial. Speeds and distances though, thank god, are metric.
All this is just kinda there and everyone's OK with it, but it is an epic mess if you think about it.
I like having more choices for units. Sometimes the "correct" unit is extremely inconvient to deal with, either because the unit sizes are oddly out of proportion with the things being measured, or the things being measured have odd ratios with the units. Sometimes even making your own unit system or going with pure ratio relationships between objects is the most useful and effective way to measure things. And I feel that people who only ever use a single system of measurement often fail to see it and put themselves at a large disadvantage.
To me using only a single system of measurement is the same as only ever using a single number base. Yeah it helps to have a standard number base everyone can use like base 10, but that doesn't mean we should try to eliminate other number bases from our vocabulary or understanding because they obviously have situational advantages.
Also from my personal bias I much prefer fractional measurements and people go apeshit if you use fractional metric units but don't blink an eye at fractional imperial or other 'non-standard' measurements.
Andrej Karpathy seems to me like a national (world) treasure.
He has the ability to explain concepts and thoughts with analogies and generalizations and interesting sayings that allow you to keep interest in what he is talking about for literally hours - in a subject that I don't know that much about. Clearly he is very smart, as is the interviewer, but he is also a fantastic communicator and does not come across as arrogant or pretentious, but really just helpful and friendly. Its quite a remarkable and amazing skillset. I'm in awe.
Agreed. I'd also add he's intellectually honest enough to not overhype what's happening just to hype whatever he's working on or appear to be a thought leader. Just very clear, pragmatic, and intellectually honest thought about the reality of things.
reply