monkeydreams

> Which brings it much closer to LLMs in this case (can an LLM talk about a topic it has never seen in training? unlikely if it cannot be derived from other training data)

This is a great experiment. Do some world building offline about Smergle the Keen finding the Hoopla of Kynashiusous. Then ask ChatGPT what reagents would have been more effective in breaking through the borantiastic door than squeeble stomach juices.

Of course we know the answer - it will either drop into a 'I have only been trained on data before blah blah blah' or 'I cannot talk about current world leaders' or 'I have not been trained on your hackneyed and unpublished romantasy saga, why don't you tell me about it?'

On the other hand, if it is aware of the general concepts of your concept, and has bridging concepts for related topics it has been trained upon, it might try to apply the general concepts to a specific but different scenario.

AI has useful features but the majority of use cases are not useful but are thrust upon us as an excuse to extract more data.

Even if the worst happens in the AI market some uses will still remain. Far fewer, I suspect, than will be demand the computational capacity of all the DCs popping up everywhere.

> LLMs are very convincing and persuasive.

The goal of LLM's, as they are marketed now, is to drive engagement and stickyness of products. A wrong answer is brushed off with a "Hey, you're right, let's try that again" - a response purposely designed to maximise the friendliness of the system and minimise the sting of a wrong answer. The fact that an LLM will not respond to the same question in the same way twice (i.e. the 'temperature' ) is because increased accuracy will not drive engagement and, therefore, increased accuracy cannot be allowed to get in the way of engagement.

Glad someone else felt the same. Discovery started well but the whole "burn" thing was... stupid. There are a million narrative devices you could use to break the ability for ships to travel via warp and they chose the least sciency of them all.

On top of this Ellison is likely to have to sell off the dried husk of Cerner in order to fuel his circular AI transactions with NVidia.

That "Smoke on the Water" MIDI tune playing through speakers at the computer shop... I cannot explain how amazing that was.

> Perhaps aesthetic - both Windows 1.0 and 2.0 were (to me at least) very ugly.

But it was amazing for those of us used to black and white/green/amber screens in DOS. You could put an image as your background. And it stayed there, lurking behind your word processor or spreadsheet, to spring back into your vision whenever you finished up your work.

I was doing tech support through the Windows 95/98/ME period and it was hell. Everything either crashed the OS or required a restart if you touched it.

When Windows 2000 rolled around and I saw how stable it was, I went out and bought it to put on my gaming PC. Another friend from work laughed at me and told me how terrible "Windows NT" was for running games until he saw how smooth Starcraft ran on it.

Yeah, Windows 95/98/ME were terrible.

The exploit of this vulnerability is so fucking trivial it defies belief. It's not a bug, it is an undocumented feature of the system.

Because I had two work calls and a child interrupt my viewing.

> It could be that simply slowing down and clearing my mind for that time would do the same.

Certainly I could not watch this in one, unbroken, session without interruption.

[flagged]

This should be a strike against MS's trustworthiness, if true. A lot of workplaces are hesitant to utilise AI models due to privacy or sensitivity concerns.

So we outsource the serial calculations to the cloud but handle branching on our cpus?

> I think people need to recognize that in many aspects what's happening is connected to societal issues that gun control and gun regulations will have very little impact on - remember, even in Japan somebody could make some kind of battery ignited home-made shotgun and kill Shinzo Abe.

... having said that, isn't it funny just how much gun violence there is in the one developed country that allows for open slather gun ownership. It's like, yes, you can never stop a determined person from doing violence, but by reducing the availability and power of fire arms you do stop a lot of fools from doing "mass shooter" levels of damage.

Selling Rams to people who do not require a huge vehicle, on the basis that the vehicle is loud and intimidating to other road users.

> other than admission to defeat compared to Chinese EVs

Or, to put it in another way; better and cheaper options for consumers.

> in every country that doesn’t have domestic car industry or huge tariffs against them.

Not every country needs a local car industry. Having a small number of efficient manufacturing countries means that everyone (both the consumers and the manufacturing countries) are better off.

As a non-American, there is nothing special or magical about the US economy besides its size. American cars are generally sub-standard and are increasingly unpopular in my country as they are either too large/loud or saddled with US politics.

> There are already cheaper, better vehicles being produced in China that American consumers cannot buy due to exhorbitant tariffs.

As an example, Australia has lifted many of the import barriers to entry to the car market and is flooded (in a good way) with cheap, reliable, and safe electric vehicles. US cars, once heavily dominating under local branding, are vanishingly rare (except in the "light truck wanker" marketplace).

> perhaps Amazon shouldn't require them in their recruitment and performance review processes

They are selecting for people who will "play the game" or, even better, will believe proactively.

No one with a lick of sense would believe that Amazon strive to be the best employer in the world. But someone who is capable of doing, for e.g., a highly skilled coding job and who believes that Amazon actually strives to be the best employer, is a rare beast who will likely not unionize at the drop of a hat.

> That hardly applies now.

I respectfully disagree given the basic fragility of the Suez Canal system along with the fact that Egypt is not growing into a major world power on the back of their governance of this canal.

Nations holding the region of Turkey are historically very powerful. My guess is that historical forces are reasserting themselves and that we may could see a much more assertive Turkey going forward.

It was confirmed by Grok so....

> but a couple of important ones

If this was the first "mistake" made by the new administration and DOGE, then these items would be understandable. Or the second mistake. Or the third, maybe.

This is intentional. That is the only conclusion one can draw now. From the terseness of the letters of dismissal to the unreliability of the message ('you're fired! Wait, no you're not!'), one can only assume this is part of the destruction of democratic institutions that the current administration is pursuing.

Gollum took back the ring after Frodo - possessed by the spirit of the ring and in its voice - cursed him with death if he should ever break his oath. Gollum did so the ring did, bringing about his downfall.

I think the problem is that the FBI are infested with these fools.

> Hallucinations are present, but usually they’re pretty minor (screwing up gender, years).

And if all hospitals were doing was having doctors treat patients, this would be ok. But healthcare is fueled by these "minor" details and this will result in delays in payment and reimbursent, trouble with patient identification, corruption of clinical coding, etc.

> considered fair use in AI training

By AI trainers, if not by the authors whose works were encoded.

Best of luck with your search. I'm not the candidate you are looking for but something about your product description makes that nice 'click' noise in my brain.

These are not exactly equal. I'm not saying that I am particularly triggered by such titles (even when they apply to my family history, genealogy, etc) but there are some people who are and accomodating them is not a huge impact.

Changing the names back because you were upset that somone changed them in the first place, with the express knowledge that some people may be affected by this, is a dick move.

One of these moves is a virtue signal, yes, but it has no real impact once completed. This current move from 'main' to 'master' is designed to both virtue signal and to upset/piss people off/etc.

This sounds like an interesting idea that has scope beyond the legal profession and into areas such as shared service management (where you might need multiple levels of management to review, edit, sign-off a comms before sending).

An 'approval to send' option might also help here.