Are they not doing music anymore for each release?

What's really unfortunate here is that due to OpenAI's drastic exaggeration of what happened here it distracts from the real capabilities here. Being able to efficiently find sources in the literature is an incredibly useful tool. And even aside from that there are now multiple examples where professional mathematicians have used GPT-5 in the thinking mode to make progress on math problems. Nothing as major as any Erdos problem, but still clear use. Terry Tao for example used GPT-5 in thinking mode to help locate a counterexample to a conjecture here https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fmathoverflow.net%2Fquestions%2F501066%2Fis-the-least-common-multiple-sequence-textlcm1-2-dots-n-a-subset-of-t. Now, he could have almost certainly done this on his own, but it clearly saved time. Similarly, computer scientist Scott Aaronson used it to get a specific useful suggestion for a function with specific properties he needed that he was then able to use to do a specific thing https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fscottaaronson.blog%2F%3Fp%3D9183. In neither of these cases was anything deeply groundbreaking done by the LLM. But the LLM clearly helped and likely saved many hours of work otherwise. And these systems continue to improve.

Star Trek is definitely not community. It is post-scarcity. The idea of some sort of "post-scarcity" society is itself pretty unlikely, but the broader ideas of a prosperous free society where advanced technology is used to help people, better ourselves and explore the universe is very different than one where drones and self-driving cars are being used by cops for unclear purposes.

This seems like this is going too far in the other direction. The news is often very negative but that's because we don't have headlines like "Last month, more solar power was installed than any other week in history, again for the 40th month in a row." Similarly, many diseases that were death sentences a few years ago have with the advance of modern drugs been turned into manageable illnesses. Cystic fibrosis for example used to kill early almost everyone who has it. Now we have drugs which make people with it likely to have lifespans close to normal. Similarly, HIV invariably lead to AIDS and a functional death sentence within a few years. Modern HIV treatments give people with HIV life expectancies better than a typical person in the 1950s. Lots of positive things are happening even as lots of negative thing are happening also.

I would much rather live in Star Trek than whatever cyberpunk dystopia we seem to be living in. Can the writers please change the genre a bit?

When I was a kid, we too had stupid things. Besides elephant jokes (how many elephants can you fit in a VW Beetle? Five -- two in the front, two in the back, and one in the glove box), the dumbest joke I remember was to run up to a friend fast and breathless and demand "Guess what!" as if you'd seen a UFO or fire engine run by, then shout "Langendorf bread, that's what!" and run away cackling like Kamala Harris.

No one said "Doctors know everything." Biology and life are weird and complicated. But anecdote are by nature not as reliable as actual studies. They lack anything like a control, and they involve small groups of people. I'm also not sure why you feel a need to reply with so much vitriol. The primary point I was making is that the literature is mixed in regards to what impact it has. How that turns into claiming that one must be "right" about somethin is beyond me. If we are going to make this personal as you seem to prefer, maybe the problem is that people who aren't very bright have trouble reading actual studies so they like to dismiss them rather than actually grapple with the evidence?

https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Freason.com%2Fvolokh%2F2025...

It's a legal blog with a legal perspective.

Of course there's no easy way to use a screenshot directory that isn't in OneDrive. That's great.

There may be an AI bubble. I'm not sure though why you think this deal is evidence that that bubble is bursting. Do you want to expand on your reasoning?

Whether marijuana helps prevent Alzheimer's or makes it worse is unclear. There are plausible mechanisms where it could do either and the empricial data is itself mixed. See https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwww.therecoveryvillage.com%2Fmarijuana-addiction%2Fmarijuana-and-dementia%2F and https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fpmc.ncbi.nlm.nih.gov%2Farticles%2FPMC7259587%2F for a start. The situation is also complicated for other forms of dementia with small amounts of marijuana use having some evidence of a slight protective effect but heavy usage showing more dementia and early cognitive decline. See https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fjamanetwork.com%2Fjournals%2Fjamaneurology%2Farticle-abstract%2F2832249 https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwww.eurekaselect.com%2Farticle%2F138726. The effects here though are small, and given pot's very high increase in heart attacks and other cardiovascular problems, any positive benefit from dementia protection is swamped by those large negatives.

1) We're going to soon have better genetic engineering anyways. 2) The vast majority of people have no problems having kids. 3) Many people who cannot have children who want them cannot have children due to reasons that have nothing to do with genetics. For example, people can have serious injuries to their genitalia. Take for example people who have been injured by landmines or in car crashes. And as emergency care has gotten better, more o those people are surviving. 4) Aside from all the practical issues, maybe let people make their own personal decisions about how to use technology instead of imposing your authoritarian aims about imagined worry which will appear centuries in the future if ever?

How gullible are you, and how gullible do you think I am, or anyone else?

They've been screaming for decades that 1.5C is the panic point, and 2.0C is right out. There's your tenths.

Most human generated code is riddled with security holes and it takes a lot of careful work to even clear out the most basic ones. AI generated code is not different. (That said, I would not be at all surprised if this code crashed frequently or had other serious problems. If it can actually turn out a genuinely useful program this way, this would be a major step forward.)

I'd be good to instead or additionally look into redesigning or banning sports that involve frequent head trauma.