Drat, that's not a very large effect of COVID vaccination. I knew that protection waned after a few months, but I expected the initial effect to be much larger.
Significant reduction in ER visits, hospitalization, and death regardless of previous COVID experiences.
October 14, 2025 at 4:55 PM
Drat, that's not a very large effect of COVID vaccination. I knew that protection waned after a few months, but I expected the initial effect to be much larger.
The brilliant, hilarious, bestselling investigation into space settlement is now in paperback! Leaving Earth and settling Mars feels tempting these days, but have we thought this through? A CITY ON MARS has answers. Learn more: www.penguinrandomhouse.com/books/639449...
LLMs make up for their lack of real-world experience with their massive exposure to the reality-modulated words humans have written. The result feels like consulting a very well-informed and helpful friend or colleague who occasionally makes human-like errors.
July 26, 2025 at 5:28 PM
LLMs make up for their lack of real-world experience with their massive exposure to the reality-modulated words humans have written. The result feels like consulting a very well-informed and helpful friend or colleague who occasionally makes human-like errors.
What we call ‘thinking’ is our brains putting words into combinations, based on the word-probability patterns it has learned. Large language models do lack our ability to modulate these probability patterns by direct experience of the physical and social world. /3
July 26, 2025 at 5:27 PM
What we call ‘thinking’ is our brains putting words into combinations, based on the word-probability patterns it has learned. Large language models do lack our ability to modulate these probability patterns by direct experience of the physical and social world. /3
We make a big mistake when we complain that large language models 'don't think like humans do'. The mistake isn't about how LLMs think, but about how humans think. Like LLMs, humans learn how to use words by experiencing how they are used by others. This is exactly what LLMs do. /2
July 26, 2025 at 5:25 PM
We make a big mistake when we complain that large language models 'don't think like humans do'. The mistake isn't about how LLMs think, but about how humans think. Like LLMs, humans learn how to use words by experiencing how they are used by others. This is exactly what LLMs do. /2
Here are the first-round arseniclife reviews and authors' response: www.usatoday.com/story/tech/c.... They're VERY credulous; the reviewers appear to have been chosen for something other than knowledge of basic microbiology.
Here are the first-round arseniclife reviews and authors' response: www.usatoday.com/story/tech/c.... They're VERY credulous; the reviewers appear to have been chosen for something other than knowledge of basic microbiology.
Here Mike Lynch mounts a fierce challenge to two recent attempts to understand the evolution of complexity in biology and beyond. I wouldn't like to place bets on the winner in this bout, but it does seem Lynch presents some tough questions. www.pnas.org/doi/10.1073/...
Here Mike Lynch mounts a fierce challenge to two recent attempts to understand the evolution of complexity in biology and beyond. I wouldn't like to place bets on the winner in this bout, but it does seem Lynch presents some tough questions. www.pnas.org/doi/10.1073/...
Nice to see the Wells' work getting due recognition. Before COVID there was nothing in Wikipedia about their work, so in May 2020 I created a page on the Wells Curve.
Thanks to @pjvogt.bsky.social for having me on the Search Engine podcast. I had a great time talking about William and Mildred Wells, two major figures in AIR-BORNE. Why they were forgotten says a lot about how science actually unfolds. www.searchengine.show/listen/searc...
Nice to see the Wells' work getting due recognition. Before COVID there was nothing in Wikipedia about their work, so in May 2020 I created a page on the Wells Curve.
The controversy poisoned her career only because she was (and remains) unable to accept the scientific community's consensus that her methods were seriously flawed and her conclusion entirely wrong. Scientists unable to admit their mistakes lose the respect of their peers.
The controversy poisoned her career only because she was (and remains) unable to accept the scientific community's consensus that her methods were seriously flawed and her conclusion entirely wrong. Scientists unable to admit their mistakes lose the respect of their peers.
🧪A child born with Cystic fibrosis in the ’50s could expect to live until age 5. In the early 2000s, age 35. With Trikafta, those who begin taking the drug in early adolescence, a recent study projected, can expect to survive to age 82.5—an essentially normal life span.
Grammarly is driving me nuts. Why does the box with the suggestion keep disappearing, and why do I have to wait so long or move the cursor around so much for it to reappear?
October 22, 2023 at 9:26 PM
Grammarly is driving me nuts. Why does the box with the suggestion keep disappearing, and why do I have to wait so long or move the cursor around so much for it to reappear?