It is hard to test things that are suspected to cause cancer because you cannot go around giving people cancer, so studies are often cut off before they become conclusive.
A small sample size is not the same as a statistically invalid size. A bad study is one that violates double blind or leaves out data or otherwise does not follow good science. A small sample size provides useful data to contribute to an overall picture. In aggregate with other studies, it adds to the picture. California adds the warning when there is evidence of a problem under some circumstances even if we do not yet know the full scope of those circumstances. It errors toward safety, which is excellent public policy.

I have wondered if we should be building a âoethoroughly debugged codeâ repository, where lots of programmers vet the code, and then see what happens with an LLM trained only on that.

Not in my experience with California. It generally means âoegood study but small sample sizeâ or âoemultiple good studies with mixed resultsâ.

Parents need to create a Google account (and probably the other FAANGs) on the day a kid is born to make sure they age into the Internet.

Not everyone. Just the teens who legit need nannies.

Calif does have data to back all of the claims. When the data is not definitive, then whether the strength of that data is sufficient to make you not use the product is left for you to decide. That's why the products are not banned. It is informed freedom.

YMMV. May have something to do with the topics I'm asking about -- they're mostly within domains having to do with technical specs where there's lots of highly replicated detail across the web, rarely about things that are niche topics, so I suspect I'm usually within domains where the LLMs get heavy training. I haven't really tried to look for a pattern in when the answers are wrong. I'll see if I notice anything about the 10%... maybe those questions are more in line with what you tend to ask about?

That may have been in the early days, but the Google summaries are improving on a weekly basis. I've been tracking how accurate they are, and it's been steadily ticking upward for the last 4 months. I'm well into 90% trustworthy at this point for anything that is wikipedia level trivia. Yeah, that's not good enough for some topics, but it's pretty amazing how solid it is, and for many topics, it is well past good enough. And often it is as good as "random website that I found says..." that I had before.

I don't like that I'm encouraging the decay of the web's business model, but I do like getting answers that are sufficiently trustworthy faster.

Also, somethings can be checked in other ways faster than following the links. For example, if I'm looking up the syntax for a library function or command line call, I'll know as soon as I try it whether the command was right or not.

The Google summary does provide links. The links are in the sidebar on desktop. On mobile, they're the little paperclip icons next to each paragraph.
With so much to rail on against AI, we don't need to hallucinate more complaints... that's the AI's job.

This is not editing. This is using DNA straight up from someone who could already have children. These are genes that are already in the population mix. Nothing new introduced. If the donor mother had had sex with the father and those children had sex with the mother, the same child could potentially result. Stop being paranoid.

How is that different from any other two people deciding to become parents by traditional means? Any time two people act to intentionally make a kid, it's a claim that they're worth existing in the next generation. That's the same narcissism.

How many children do you think these people are going to have??? "Bring down the species" level of problem would require a huge number of other humans to die off to reduce the population down to where these are at all statistically significant. And even if the tech spreads beyond this one experiment, it's not going to cover the majority of even first world nations that could afford the operation.
And, let's say there is a bad combo, then we can switch in another set of mitochondrial DNA for the next generation.

No, this could not bring down the species. That isn't even remotely among the possible results of this experiment.

But dependency on Internet is a service situation we choose to buy into when we buy the console product. That is not an unexpected situation, unlike a surprise stoppage when a user thinks a game is legit. Anyone who does not like Internet dependency should not be buying the consoles period. Games are not required for life, so I do not have sympathy for anyone who buys into that system and then complains about the base model. Plenty of games outside the console world do not have that limitation.

Yes, agreed. But Nintendo does not know which is which, so they ding both accounts and then, per TFA, restore one after further evidence. How is that wrong?