blurbleblurble

3 ms·

1.
by blurbleblurble 2mo ago
You'd think they'd be able to read the room with all that artificial intelligence. This is just horrible PR. What use is the product if you're just cannibalizing your prospective market?
2.
by blurbleblurble 2mo ago
Existential
3.
by blurbleblurble 2mo ago
https://youtu.be/k_gaZjXD5OY?t=200
4.
by blurbleblurble 2mo ago
Because everything is converging on a backlog of huge efficiency gains established in research, waiting to be combined. Looped transformers, a whole host of diffusion techniques and new quantization techniques, maturation of ternary distill
5.
by blurbleblurble 2mo ago
It's going to happen very soon, which is why these frontier labs are scrambling to shut down open source language models. There's an existential risk threatening their obscene returns.
6.
by blurbleblurble 2mo ago
The beginning of the bubble pop?
7.
by blurbleblurble 2mo ago
Or maybe even to run parallel threads of the same model!
8.
by blurbleblurble 2mo ago
They should have waited until after they fixed Opus to do this.
9.
by blurbleblurble 2mo ago
open-jarvis seems cool too
10.
by blurbleblurble 2mo ago
Or a lot worse
11.
by blurbleblurble 2mo ago
Such bad PR for OpenAI! Who approved this?
12.
by blurbleblurble 2mo ago
Everyone's talking about flock but what about the IEEE ISAC and the broader 5G/6G projects? They're running laps around flock on surveillance. https://www.youtube.com/watch?v=SMox8qeMvlY
13.
by blurbleblurble 2mo ago
> the anti-distributional semantics crowd Who is this crowd specifically? The stochastic parrots people? Noam Chomsky? I don't think they're good representatives of media theory at all whatsoever. The humanities are much more d
14.
by blurbleblurble 2mo ago
I think you're slightly misunderstanding my point, which is that a huge spectrum of associative pattern seeking logics are embedded in language and that the LLM learns them and operates them, approximately. "There is nothing to su
15.
by blurbleblurble 2mo ago
"we demonstrate that LLMs can spontaneously develop novel social biases about artificial demographic groups even when no inherent differences exist" It's almost as though bias-making machinery is embedded in the texts these t
16.
by blurbleblurble 2mo ago
Same attitude that prevents the urinal manufacturers from switching I reckon
17.
by blurbleblurble 2mo ago
> strict vegan diets sabotage achieved
18.
Billionaires Are Funding 'Anti AI' Content [video](youtube.com)
4 points by blurbleblurble 2mo ago|1 comments
19.
by blurbleblurble 2mo ago
Absolutely agree. Perverse incentives are at play propped up by the the big lie that these tools somehow are magically separate from us. They are not. There's an accountability gap right now that's fueling resentment ripe for misd
20.
by blurbleblurble 2mo ago
Burnout is learned and/or imposed helplessness. When your agency is missing it's like part of you is dying.
21.
by blurbleblurble 2mo ago
Friends don't let friends use SMS/MMS
22.
by blurbleblurble 3mo ago
If it's not watermarking then what is it?
23.
by blurbleblurble 3mo ago
gg
24.
by blurbleblurble 3mo ago
Cause they're a bot
25.
by blurbleblurble 3mo ago
Reasonable suspicion of what though? Is them just being suspicious by default "reasonable"?
26.
by blurbleblurble 3mo ago
Evidence of what?
27.
by blurbleblurble 3mo ago
If the law isn't hackable what do you call what these billionaires and religious fundamentalists are doing to it right now?
28.
by blurbleblurble 3mo ago
Time for memristors?
29.
by blurbleblurble 3mo ago
I'm interested to understand your theory better but right now the way you've expressed it is hard to make sense of for me.
30.
by blurbleblurble 3mo ago
Since when was surveillance disconnected from civil rights? Since never.

More ›