Yes. It’s crazy. That’s why the vast majority of us don’t do it.
It’s one thing to be a vegetarian for health or environmental reasons.
When you try to convince people that meat==murder, you come across as a wackadoodle.

That’s great news. The other 9 of the 10 biggest protests were were extremely successful at affecting change.

Since we made such massive progress on all the others, this is clearly a harbinger of social and political progress.

I wouldn’t either but that’s exactly what lmsys.org found.

That blog post had ratings between 858 and 1169. Those are slightly higher than the average rating of human users on popular chess sites. Their latest leaderboard shows them doing even better.

https://lmarena.ai/leaderboard has one of the Gemini models with a rating of 1470. That’s pretty good.

I imagine the “author” did something like, “Search http://google.scholar.com/ find a publication where AI failed at something and write a paragraph about it.”

It’s not even as bad as the article claims.

Atari isn’t great at chess. https://chess.stackexchange.com/questions/24952/how-strong-is-each-level-of-atari-2600s-video-chess
Random LLMs were nearly as good 2 years ago. https://lmsys.org/blog/2023-05-03-arena/
LLMs that are actually trained for chess have done much better. https://arxiv.org/abs/2501.17186

Like humans are way better at answering stuff when it’s a collaboration of more than one person. I suspect the same is true of LLMs.

It is.

It’s really common for non-language implementations of neural networks. If you have an NN that’s right some percentage of the time, you can often run it through a bunch of copies of the NNs and take the average and that average is correct a higher percentage of the time.

Aider is an open source AI coding assistant that lets you use one model to plan the coding and a second one to do the actual coding. It works better than doing it in a single pass, even if you assign the the same model to planing and coding.

The Israelis regularly murder journalists and civilians. The danger was quite real.

Sometimes it seems like most of these AI articles are written by AIs with bad prompts.

Human journalists would hopefully do a little research. A quick search would reveal that researches have been publishing about this for over a year so there’s no need to sensationalize it. Perhaps the human journalist could have spent a little time talking about why LLMs are bad at chess and how researchers are approaching the problem.

LLMs on the other hand, are very good at producing clickbait articles with low information content.

That makes sense. Not everything needs to be testable. There are many interesting and important ideas outside of science.

The main problem would be if someone wanted to set policy based on it. That includes the implicit experiment of, “If we adopt policy A we can expect outcome B.” If we haven’t tested that before turning it into a policy, the policy itself becomes the experiment, and then we need to be very careful about the ethics surrounding such an experiment.

It’s kind of like string theory. It has a bunch of interesting conjectures but nobody can figure out a way to test any of it.

Take the “selfish gene” (the idea predates Dawkins). One of the theories states that it may be evolutionarily advantageous for an individual to sacrifice themselves for the group if they share enough DNA. They lose the DNA in their bodies but save the exact same DNA in the bodies of their extended family. That’s a nice idea and you can get the math to work out in game theory models but how do we test if that’s why ducks sometimes lag behind when a hunter tries to shoot them?

That’s not to say it can never be tested. There are other cases where we needed to wait for technological breakthroughs until theories could actually be tested.

It’s 27T Pro. I like it better than the iPhone it replaced.

The only downsides I’ve seen so far are that it requires a separate app for wifi calling and it has fewer zoom options for the camera. I’d like to figure out how to get the IR blaster to read signals (so I can easily clone my remotes).

Yeah. I’m typing this on a $300 Chinese phone with 10600mAH battery, reverse wireless charging, a thermal imaging camera, and it’s waterproof and shock resistant.

You can add lots of things. I tend to throw in liberal amounts of smashed garlic and some mustard. Depending on what’s available in the garden, I may throw in some fresh herbs. Sometimes I toss in a little lemon zest or a finely smashed caper.

But none of that is needed in a simple vinaigrette.

Mustard will make it better but you don’t need it.
You won’t get as good an emulsion and it will separate faster. Once you pour it on some salad it will be pretty hard to notice that.

If someone is at the point in their cooking journey where they’re asking how to make vinaigrette, I keep it as simple as possible. TBH even the pepper isn’t strictly necessary. Many people don’t have pepper grinders and preground pepper doesn’t add much flavor.

Salt is the only one that I’d say is absolutely necessary.

It’s a bit complicated for a “simple vinaigrette”.

Pour about equal amounts of oil and vinegar in a jar; add a little salt and pepper.
Screw on the lid and shake that shit.
Pour it on stuff.

Sauces can get arbitrarily complicated. If someone wants a simple recipe, keep it simple.

Germany tried to create laws to prevent a repeat of the Holocaust. It’s a laudable effort but they’re failing at it.

The problem is that they were so specific about preventing “The Holocaust” that they ignored many other kinds of bigotry and racism. They thought that if they forbid a few key phrases and symbols, hatred would wither on the vine. Instead they just cleared the way for other aspects of racism to flourish.

talk to their parents first

This is key. It’s one thing to stop a child from harassing your pet. If you “teach them a lesson” after the fact without the parents’ permission, there are now two wrongs that need to be dealt with.

Exactly. The real debate is on which parts should be off limits.

Most people can think of some speech that they consider so horrible that nobody should be allowed to say it.

People often try to hedge that position by arguing that they’re not even really infringing on anyone’s speech because their form of restriction doesn’t meet a sufficient threshold of censorship.

Does anyone?

The closest I can think of to “real free speech absolutists” is the old-school doctrinal libertarians. Even they have limits on what they believe should be allowed and specifically state that contracts should be legally enforceable.

There is already a foolproof method that is immune to any abuse of trust by admins; create an alt account.

Yeah. And the fix for that has nothing to do with “de-duping” as a database operation either.

The main components would probably be:

1. Decide on a new scheme (with more digits)
2. Create a mapping from the old scheme to the new scheme. (that’s where existing duplicates would get removed)
3. Let people use both during some transition period, after which the old one isn’t valid any more.
4. Decide when you’re going to stop issuing old SSNs and only issue new ones to people born after some date.

There’s a lot of complication in each of those steps but none of them are particularly dependant on “de-duped” databases.