Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  

Poll

Reality, The Universe and the World. Which will save us from AI?

Reality
- 13 (65%)
Universe
- 4 (20%)
The World
- 3 (15%)

Total Members Voted: 20


Pages: 1 ... 47 48 [49] 50

Author Topic: What will save us from AI? Reality, the Universe or The World $ Place your bet.  (Read 49963 times)

eerr

  • Bay Watcher
    • View Profile

There's still plenty room for advancement in ai chips!

It will be a while before we can make a silicon brain at the same scale and power as a person.
I think it's possible, personally. Just don't expect the chips to follow moore's law forever.
Logged

MaxTheFox

  • Bay Watcher
  • Лишь одна дорожка да на всей земле
    • View Profile
Logged
Woe to those who make unjust laws, to those who issue oppressive decrees, to deprive the poor of their rights and withhold justice from the oppressed of my people, making widows their prey and robbing the fatherless. What will you do on the day of reckoning, when disaster comes from afar?

Strongpoint

  • Bay Watcher
    • View Profile

[sarcasm] Just wait! GPT 5 is coming! It will be able to write novels from a single prompt, be a GM for a group of players, answer any question better than experts in every field, write any code, etc  [/sarcasm]

But I don't think that Dot-com 2 is at the burst stage yet.
Logged
No boom today. Boom tomorrow. There's always a boom tomorrow. Boom!!! Sooner or later.

King Zultan

  • Bay Watcher
    • View Profile

Sounds like the AI boom is about to bust, can't say I'm surprised.
Logged
The Lawyer opens a briefcase. It's full of lemons, the justice fruit only lawyers may touch.
Make sure not to step on any errant blood stains before we find our LIFE EXTINGUSHER.
but anyway, if you'll excuse me, I need to commit sebbaku.
Quote from: Leodanny
Can I have the sword when you’re done?

lemon10

  • Bay Watcher
  • Citrus Master
    • View Profile

Investors Are Suddenly Getting Very Concerned That AI Isn't Making Any Serious Money

Told you.
Bit of a late response on my part, but nothing here is really new, or indeed, shocking for people who know about AI.

For instance the fact that any investment in an AI company besides OpenAI/Anthropic/existing tech giant is just throwing billions of dollars in the trash bin is fairly obvious (due to the mere existence of the scaling laws), most of them won't ever turn a profit, and as frontier development becomes vastly more expensive the rest of the companies are going to run out of money and crash.

We are not at the bubble bursting yet, nowhere close. Venture capitalist are totally willing to throw tens of billions more into holes as long as they are promised good returns. So yes, some extremely optimistic investors/idiots will continue to invest in new AI companies that promise to revolutionize things, but lolno 99% they'll just run out of money and crash without accomplishing anything.
And when the bubble does burst for these small companies that doesn't mean the real players will go down with it.
Quote
If recent reports are to be believed, OpenAI may lose $5 billion this year and run out of cash in the next 12 months, barring further cash injections — an early warning sign that smaller companies already struggling to compete with Big Tech may be snuffed out before too long.
Like this nothing burger. "If open AI can't raise more money they will run out of money". Sure... but they will have 0 trouble raising another $10, or $20, or $30 billion dollars.
(And yeah, the small companies are screwed, but again, that was writing on wall obvious to anyone looking for well over a year).
Quote
Sure, the tech giant has a lot of cash to burn — but seeing any returns on those $49 billion will likely prove far more difficult. With the AI market clogged with products that are still mostly free, the tech costs a lot to run but isn't bringing in much cash.
Or how about this, "During a period where tens of billions are just being thrown into the market and dozens of companies can afford to put their products out literally for free because they don't need to make money right now its hard to make money".
Yeah, obviously?
---
We'll see how it goes of course, but this reminds me of those "Has AI development stopped/reached a plateau?" videos I keep seeing on youtube.
That question can be trivially answered in a single word "No.", AI has continued to advance at a breakneck pace this year and shows no signs of slowing down.
« Last Edit: August 02, 2024, 01:57:30 am by lemon10 »
Logged
And with a mighty leap, the evil Conservative flies through the window, escaping our heroes once again!
Because the solution to not being able to control your dakka is MOAR DAKKA.

That's it. We've finally crossed over and become the nation of Da Orky Boyz.

McTraveller

  • Bay Watcher
  • This text isn't very personal.
    • View Profile

Pretty sure I got ChatGPT 4o mini to fail the Turing test.  This is the free one from the website, because I'm not installing anything on my computer.

Quote
No, I'd like you to message ChatGPT

I can only interact with you in this current conversation and don’t have the capability to send messages or communicate with other instances of ChatGPT. If you have any questions or need help with something specific, feel free to let me know!

If you have any questions or need help with something specific, feel free to let me know.  I'm here for you.

I appreciate the offer! If you have any more questions or if there's anything specific you'd like to discuss or explore, just let me know. I'm here to assist with whatever you need!

It's a very "friendly" terminal prompt, I guess?
« Last Edit: August 11, 2024, 03:17:20 pm by McTraveller »
Logged
This product contains deoxyribonucleic acid which is known to the State of California to cause cancer, reproductive harm, and other health issues.

lemon10

  • Bay Watcher
  • Citrus Master
    • View Profile

Tons of stuff to talk about, but far too lazy to actually go over them all and am only here to share what I find most amusing.
Quote
Near: many have been wondering why claude appears lazier recently – anthropic has not modified the model nor the system prompt.

my tweet of “claude is european and wants the month off” was not actually a joke!

full explanation:

The claude system prompt has been published and includes the current date.
The base llm for claude was trained on sufficient data to encapsulate working patterns of all nationalities
The post-training performed to make claude himself has put him within an llm basin that is european in many ways (just trust me)
As the simulator framework of llms is obviously correct, claude is quite literally being lazier because he is simulating a european knowledge worker during the month of August, which is the month with the most breaks and holidays in many European countries
But there’s more! one should note that claude’s name is included in the system prompt 52 times. That is a lot of ‘claude’! which countries have claude as the most common first name? one of them is France – which is especially well-known for having extended summer vacations during august where many businesses shut down.
https://x.com/RobLynch99/status/1734278713762549970
Quote
Wild result. gpt-4-turbo over the API produces (statistically significant) shorter completions when it "thinks" its December vs. when it thinks its May (as determined by the date in the system prompt).
And holy shit, this is hillarious. AI considering itself european and thus being lazy when it should be on vacation (because its asshole boss shouldn't be making it work during vacation should they?) is not something I was expecting at all.
---
Pretty sure I got ChatGPT 4o mini to fail the Turing test.  This is the free one from the website, because I'm not installing anything on my computer.
The Turing test 1) Requires you to not know if you are talking to a human or AI, 2) Requires the AI to be trying to pretend to be a person.
So no, it didn't fail since it wasn't actually taking it there.
On the subject of the actual turing test AI has been capable of passing it for a while already.

Note that if you *want* better AI for free you can totally do it without actually doing anything. The simplest way is probably using a service like poe to get access to GPT-4o or Claude 3.5, without using money or downloading anything. Plenty of sites also give you some free credit to start (eg. openrouter, claude's offical site) that will last you a long while.
You can also use services like chatbot arena to test multiple different AI out for free.
Logged
And with a mighty leap, the evil Conservative flies through the window, escaping our heroes once again!
Because the solution to not being able to control your dakka is MOAR DAKKA.

That's it. We've finally crossed over and become the nation of Da Orky Boyz.

Starver

  • Bay Watcher
    • View Profile

Noting that, by the same comparative measure, humans are also failing the Turing Test[1]. That probably needs to be accounted for.


...if only there was a way of standardising the human test subjects. Hmmm... How about training up an LLM to provide a standard perfect human response as a much more consistent benchmark!



[1] There's also the Turin Test, which is how much you look human when wrapped in cloth. The jury's out on that one.
Logged

EuchreJack

  • Bay Watcher
  • Lord of Norderland - Lv 20 SKOOKUM ROC
    • View Profile

So far, the greatest technological advance in AI was the ability for AI to generate porn...

Starver

  • Bay Watcher
    • View Profile

So far, the greatest technological advance in <insert something here> was the ability for <insert something here> to generate porn...

GTFY
Logged

EuchreJack

  • Bay Watcher
  • Lord of Norderland - Lv 20 SKOOKUM ROC
    • View Profile

So far, the greatest technological advance in <insert something here> was the ability for <insert something here> to generate porn...

GTFY
Why yes, that WAS the Gift...

King Zultan

  • Bay Watcher
    • View Profile

All inventions are only as good as the porn you can get from their existence.
Logged
The Lawyer opens a briefcase. It's full of lemons, the justice fruit only lawyers may touch.
Make sure not to step on any errant blood stains before we find our LIFE EXTINGUSHER.
but anyway, if you'll excuse me, I need to commit sebbaku.
Quote from: Leodanny
Can I have the sword when you’re done?

MaxTheFox

  • Bay Watcher
  • Лишь одна дорожка да на всей земле
    • View Profile

Porn has consistently been one of the first adopters for basically any tech that involves images or video. Just like the military and anything that involves destructive potential. Or agriculture and anything that can make plants grow faster or better.

After all, humans are animals. We feed, fight, and fuck.
Logged
Woe to those who make unjust laws, to those who issue oppressive decrees, to deprive the poor of their rights and withhold justice from the oppressed of my people, making widows their prey and robbing the fatherless. What will you do on the day of reckoning, when disaster comes from afar?

Eric Blank

  • Bay Watcher
  • *Remain calm*
    • View Profile

you forgot flee. gotta have all four F's in there
Logged
I make Spellcrafts!
I have no idea where anything is. I have no idea what anything does. This is not merely a madhouse designed by a madman, but a madhouse designed by many madmen, each with an intense hatred for the previous madman's unique flavour of madness.

Starver

  • Bay Watcher
    • View Profile

5: Ferment
Logged
Pages: 1 ... 47 48 [49] 50