Post reply

Name
Email
Subject
Message icon

Other options

Return to this topic
Don't use smileys

Verification:

Please leave this box empty:

Shortcuts: ALT+S post or ALT+P preview

Topic summary

Posted by "the end is nigh" clowns

- Today at 08:34:51

Funny:

Quote from: reddit.com/r/LocalLLaMA/comments/1x52lsb/new_ai_fake_text_generator_may_be_too_dangerous/"New AI fake text generator may be too dangerous to release, say creators" (2019 Guardian headline)

"OpenAI ... says its new AI model, called GPT2 is so good and the risk of malicious use so high that it is breaking from its normal practice of releasing the full research to the public in order to allow more time to discuss the ramifications of the technological breakthrough."

[..]

By today's local AI model standards this is not even laughable and we are still here.

Which models are we talking about? For up to 40B, see e.g.:
artificialanalysis.ai/models/open-source/small#artificial-analysis-intelligence-index

More about running AI models locally (and privately): notebookchat.com/index.php?topic=315954.0 ("Your own ChatGPT, offline: AI without the cloud on your laptop") and comments.

These ~30B parameters models, that you can run locally on 32 GB RAM + 8 GB VRAM, are no joke:

Quote from: reddit.com/r/LocalLLaMA/comments/1x4cm73/once_again_anthropic_doing_free_marketing_for/Once again anthropic doing free marketing for open weight models

Models of 30B parameters and under can already complete many of the individual steps for replication today, though they make mistakes when composing them end to end. In our simplified evaluations, GLM-5.2 and stronger models are capable of end to end replication. If small models improve at their historical rate, they'll match this capability level by the end of this year. Small weight-based replicators may be possible soon after.

And the top rated comment from that reddit post is:

Quotethey talking about our goat 27b

(and reddit.com/r/LocalLLaMA/comments/1x4ii8e/anthropic_is_doing_free_publicity_for_glm_again)

PS: Qwen4-27B is rumored to be released by the end of october.

Posted by jessierosman

- October 04, 2026, 20:57:01

I think the most interesting part is the difference between human extinction and other serious consequences of AI. A 5 to 10 percent chance of actual extinction sounds extremely high, but the article also makes it clear that things like misuse, concentration of power, destabilization and loss of human control could be much more likely. I also like that the author asked several different AI systems instead of relying on just one answer. The numbers should not be treated as scientific predictions, but the fact that the systems give such different estimates shows how uncertain this whole question still is.

Posted by But is AI profitable tho?

- September 29, 2026, 20:48:59

Approximately 100 years ago people thought that in 100 years humans would have wings any be able to fly (there's an picture).
At tome point humans thought that driving at faster than like 20-30 miles per hour would be lethal.
Many movie classics in the 70s to 90s imagined we'd live in a much more advanced civilization by 2020 or even earlier.

Just a few examples. Obviously, none of this happened. It seems the human psyche likes to overestimate by a lot.

So, if I was asked the same question, I'd say: "Of course not, 0.0% chance. (It's maybe even 0% chance for ever, despite that obviously no one can predict the future, or at east for a very long time.)" (but it doesn't mean that AI may still take your job away (which you probably don't like anyway, but it's still better than having none)).

In the meantime: isaiprofitable.com

Posted by duh

- September 28, 2026, 10:51:16

It is funny how someone and we know who is in charge of selling and promotion immature technology aka "thin air" state it as VERY ADVANCED AND ETC but in reality even ppl which someone we could call "Advanced" dont buy it. In other hand this bubble move stagnate economy, that is why many politicians support it, but outcome of this move yet another 2008 crisis even mote than that...

Posted by Redaktion

- September 25, 2026, 19:50:06

In the "AI asked" series, I ask various AI chatbots questions about relevant issues of our time and ask them for their own assessment. Today, we're looking at a widely discussed statement by an Anthropic employee who said AI could wipe out humanity within this decade. Here's what the AI has to say.

https://www.notebookcheck.net/AI-asked-What-are-the-odds-that-AI-will-wipe-out-humanity-this-decade.1408390.0.html