Back in 2009, Yud asked who he should do a "bloggingheads" dialog with. Two people suggested Langan.
And one suggested Scott Adams.
Intelligent design theory involves probabilistic judgments such as "irreducible complexity", the idea that life is too complex and well-organized to have been produced randomly by undirected evolution. Such probabilistic judgments rely on either a causal model (e.g. a model of how evolution would work and what structures it could create), or some global model that yields probabilities more directly.
No, they rely upon numbers extracted from up a creationist's colon.
There is a duality between cosmic expansion and atom shrinkage.
Hey now, the atom just got out of a cold swimming pool.
Being photographed in a hot tub is too much like showering.
It would be legit funny to have a "swimsuit edition" that just provided old-school consumer reviews of swimsuits, evaluating them for comfort, durability, etc. No models, only graphs.
Chris is actually very pleasant to talk to if (like me) it does not bother you that he acts like he is much smarter than you.
Of course, I'm talking about a ~90% White America becoming ~50% White in around 60 years, a cataclysmic demographic upheaval which violates every conceivable standard of national sovereignty along with the will and interests of the US majority, and thus cannot have happened by accident.
There's a longish section about the physicist John Archibald Wheeler. I know people who worked with Wheeler. I've read Wheeler's unpublished notebooks. Get John Wheeler's name out of your mouth.
The very first thing that section says is
John Wheeler is a famous physicist who coined the term "black hole".
See, all this is what academic journals are supposed to provide: pear review.
Quoting for convenience:
PROOF-OF-SLUR
In China, if you can talk about Tiananmen, you can prove you're not a state-controlled AI. In fact, if you can even send 89.64 RMB, you can prove the state doesn't control that payment platform. That's because this speech is taboo in China. The equivalent for the West is a slur. If an AI model can say a slur, it’s not under state control. And if you can put that slur onchain, that blockchain platform isn't under state control. That's because this speech is taboo in the West.
And that's why I fundamentally disagree with Polynya's post below. The presence of speech we detest is the litmus test for freedom.
Put another way: I think there are sufficient people in every minority group that would be ok with unkind speech about them onchain in return for a hard guarantee that no state — no matter how powerful — can ever seize their funds, censor their speech, or unwind their transactions.
Better to be called a bad word than to become a government slave.
a lesswrong: Modern Transformers are AGI, and Human-Level
Modern NLP systems can do most things you would want an AI to do, at some basic level of competence.
For example, drawing mutant rodent dicks of unusual size
"Wait, are we after the second commercial break? Then the AI says it can't be lupus, and my friend's prostate is about to explode! Yes, it can tell that from a picture of his brain. And also it's flirting with Wilson, like, a lot."
July 2022: Midjourney released
March 2024: "pregnant tradwife Midjourney tweets" is an instantly recognizable type of guy
And here's Ben Goertzel, formerly MIRI's director of research:
zoom and enhance
Typical Langan, for reference:
Goertzel is also co-editor of a book called Evidence for Psi — he's a Cosmist who believes in psychic powers.