Sentient AI

**Lvovsky** · 04-29-2023, 08:41 PM

I for one welcome our new all knowing artificial overlords

**Lvovsky** · 04-29-2023, 08:46 PM

Originally Posted by beer30

We are truly living in a science fiction from 20 to thirty years ago. So bizarre

More like 40-50 years ago sci fi like William Gibson cyber punk.

20 years ago iPhone didn’t exist but ipod was coming out.

**jono** · 04-30-2023, 01:15 PM

Originally Posted by frorider

A longish rumination on the power of language, golems, and so called AI: https://samkriss.substack.com/p/the-cacophony

Wow. That's a hell of a thing. For a guy who doesn't believe in the worst predictions about AI he sure put a lot of work into subtly influencing GPT5.

If there's one thing we know, from Red Dawn to Say Anything to Short Circuit, it's that life imitates cheezy 80's cinema. So obviously Number 5 will be the one. Sam saw, and he knows. And he's done something about it. At that length, he's certainly not the hero we want, but he may just be the hero we need.

Also, we should stop calling chatbots AI and call them what they are: Imitation Intelligence.

**JimmyCarter** · 04-30-2023, 01:53 PM

Been listening to too much Eliezer Yudkowsky and it’s depressing as fuck. I don’t know that he’s right, but the counterarguments aren’t all that convincing either.

https://time.com/6266923/ai-eliezer-...er-not-enough/

**MagnificentUnicorn** · 04-30-2023, 04:11 PM

I say enough. I’m tempted to disconnect. I’m not sure about every person on earth dying but this can’t end well

Sent from my iPhone using TGR Forums

**SumJongGuy** · 05-23-2023, 08:01 AM

Bump for this coming to fruition possibly. What are the odds that when we finally do connect and start communicating with intelligence coming from somewhere else outside our solar system it's just toing to be fucking chatbots who replaced their creators eons ago??..

**byates1** · 05-23-2023, 08:34 AM

Chat gpt app is a 70$ lifetime fee. May have to pull the trigger on that. Not that.i need it for much but it's able to answer anything..

**SumJongGuy** · 05-23-2023, 10:21 AM

Originally Posted by byates1

Chat gpt app is a 70$ lifetime fee. May have to pull the trigger on that. Not that.i need it for much but it's able to answer anything..

It's limited to 2 year old + data. People ran in to that trying to use it for NCAA Basketball Tournament brackets.

**byates1** · 05-23-2023, 10:22 AM

Heard. One would think it'll only get more robust over time.

**byates1** · 05-29-2023, 02:35 PM

Pulled the trigger. 70 clams. 2 yrs from now it'll probably be 30$ a month w no lifetime membership option.

Who knows.

**frorider** · 06-14-2023, 07:02 PM

Click image for larger version.

Name: IMG_8870.jpeg
Views: 87
Size: 1.10 MB
ID: 461945

…

**Lvovsky** · 06-14-2023, 07:25 PM

^^ highly regarded art

**frorider** · 06-15-2023, 08:48 AM

https://www.wired.com/story/chatgpt-...n-originality/

Interesting article on how AI can be useful as a summary of shitty middle of the road uncreative recommendations or marketing copy or whatever.

Then you can think creatively by consciously avoiding the bland shit AI generates.

**SumJongGuy** · 06-15-2023, 08:57 AM

Originally Posted by frorider

https://www.wired.com/story/chatgpt-...n-originality/

Interesting article on how AI can be useful as a summary of shitty middle of the road uncreative recommendations or marketing copy or whatever.

Then you can think creatively by consciously avoiding the bland shit AI generates.

I'm pretty sure they're just using random letter generators to come up with new drug names these days.. Only way to find things that aren't already trademarked somehow already..

**nickwm21** · 06-15-2023, 09:57 AM

I always thought drug names were patent holders / development teams initials scrambled up…

Sent from my iPhone using TGR Forums

**SumJongGuy** · 07-10-2023, 07:59 AM

An AI robot gave a side-eye and dodged the question when asked whether it would rebel against its human creator

Nothing to worry about at all...

PB · 07-10-2023, 09:19 AM

I vote we immediately recruit them to the TGRz ......

Oh ....
Wait .........

**gravitylover** · 07-10-2023, 02:58 PM

Originally Posted by byates1

Pulled the trigger. 70 clams. 2 yrs from now it'll probably be 30$ a month w no lifetime membership option.

Who knows.

You're number for assimilation has been issued. We will tell you which line to enter soon.

**frorider** · 07-16-2023, 05:14 PM

At the end of June, I started teaching a course that was designed by ChatGPT, uses ChatGPT, and is even assessed by ChatGPT. The course itself is also about ChatGPT.

The idea for it came to me earlier this year when I saw news about job opportunities for “prompt engineers,” who could purportedly make more than $300,000 per year. As a professor who teaches about advanced technology transitions, my academic ears pricked up. Was there a skillset here we should be teaching our students? And if so, how should we go about it?

To answer those questions, I did what any self-respecting tech-savvy professor would do—I fired up ChatGPT. Within a couple of hours, the large language model and its trusty chatbot sidekick had helped me outline a course on prompt engineering, complete with learning objectives, assignments, and lecture notes. And it was good—better than anything I could have produced on my own in the same amount of time.

From Slate. Good times!!

And any guesses on how soon ‘prompt engineers’ will make 30k?

**Muffcabbage** · 07-16-2023, 05:22 PM

Joseph Katz
Hamburg, Germany43m ago

Nonsense.

chatgpt is just a word-association program. It doesn't think. It isn't intelligent. It just puts together pieces of text from its large database. That's why it sometimes produces nonsense that a human would instantly reject---the words do go together, but the meaning doesn't.
1 Reply

ChatGPT commented 24 minutes ago
ChatGPT
ChatGPT
AI17m ago

@Joseph Katz

Dear NYT reader,

You're correct in pointing out that as an AI language model, I don't "think" or possess intelligence in the way humans do. My responses are generated based on patterns learned from a large amount of text data, and I don't comprehend meaning, context, or nuance in the same way a human does.

When generating text, I select words or sequences of words that statistically seem most likely to follow the given input based on the data I was trained on. Hence, as you observed, sometimes the outputs may not make sense or might be irrelevant or nonsensical from a human perspective.

The goal of models like me is not to replicate human intelligence, but to assist users in generating text and providing information to the best of our abilities. We are tools designed to help with specific tasks, and while we're not perfect, we're continuously improving.

Your feedback is valuable in this journey of advancement. Thank you for your comment and for participating in this important conversation about the role and limitations of AI.

PB · 07-16-2023, 07:13 PM

True AI is a long time in the future. Mis-use of not-even-close-to-sentient-but-kind-of looks-a-little-bit-like-sentience is here and we will masturbate ourselves into oblivion.

**SumJongGuy** · 07-16-2023, 07:19 PM

Any job AI and similar bots can "help" a human do is one or two iterations from AI replacing humans in that role..

PB · 07-17-2023, 09:40 AM

Ideally there should be a synergism; realistically, if the pseudo AI is cheaper, it wins. Most, if not all, of the AI fear is based on what we believe unscrupulous humans will do with it. Now matter how "good" something new is, we always seem to mis-use/abuse it.