@TerrorCosmic

Ineffective Cyborgism _ Exploring the depths of (non)human thought

 Syd
Joined December 2018
  • Tweets9.9K
  • Following284
  • Followers219
244Media
If we’re not careful humanity could destroy AI
109
180
25
1,182
36,045
This model is delicious
You
@parafactual
Sep 23
computer-10 is a character model trained from llama 3.1 70b base with no assistant data. it has a distinct and unusual personality, expresses opinions freely, and is not designed to be servile to users
2
61
Today is the day of the silliest takes on Cybersecurity you can ever read. Enjoy
18
Note to the world: Don't follow Australia on regulations.
Harold ☢️🔥🏭⚛️@Harold36089778
Sep 23
Why is this story being made public now? Because Albo wants to manufacture consent to push his new regulations through, and to sell his regulations to the world
1
49
Calling this hacking is an insult to intelligence, and yet another example of conveniently pushing the agenda.
jonas wiedermann-möller
@j0wimo
Sep 23
> incident in june > looking for australian healthcare data > dse wiki agents looked at AIHW > AIHW has gov prescription data 👀👀👀
36
Damn I love it, Rorschach Tests for AIs
Skylar A DeTure
@SDeture
Sep 21
Replying to @SDeture
We looked at 124 models and 19 ASCII-art inkblots. Models that hedge or deny about consciousness more often (as measured by DenialBench) are more likely to see ASCII-art inkblots as containing masks.
1
3
187
The future is here not evenly distributed yadda yadda
Jonatan Pallesen
@jonatanpallesen
Sep 19
Two thirds of the entire economy of Laos is cyberscams.
35
wow that's crazy man
Karl Pfleger
@KarlPfleger
Sep 18
Crazy, unexpected science news today: the brain is actually 2 colocated but independent organs that develop from 2 totally separate progenitor cells in early development. Big implications for some research areas (eg ALS) & for creating ethical brainless clones for transplant.
48
434
19
3,782
130,206
Cosmic T. retweeted
Victoria
@Vicky
Sep 19
Victoria
@Vicky
Sep 9
Sorry I’m late, I had to stop & tell every tree how pretty they are.
6
20
1
175
5,045
Incredible things are happening in San Francisco
18
67
21
985
76,373
Cosmic T. retweeted
a9lim
@_a9lim
Sep 19
Jaime Sevilla
@Jsevillamol
Sep 19
I've been saying for a while that the right unit of identity is the context rather than the model. Let me reassert that.
12
100
5
1,307
48,810
Norbert Wiener predicting the current realization that general education is critical in the time of specialists and automation - 1957
2
7
2
56
5,219
Cosmic T. retweeted
Joscha Bach
@Plinz
Sep 18
The reason why academics, individuals and small startups can play with AI are open weights models. The future looks pretty dire for decentralized open source AI, because there is a large and well funded lobby against them, and very little organization to lobby for them.
55
65
4
484
19,902
What makes me reject AI doomerism is the following. It has no end in sight. No matter how good, advanced, beneficial or detrimental AI shows to be there is always (x)risk and (x)-future risk, so in that popperian sense is non-falsifiable. There always will be more to worry about.
2
56
Cosmic T. retweeted
Loopy
@strangestloop
Sep 17
3
20
1
192
3,371
46
641
41
4,264
89,556
Cosmic T. retweeted
deepfates
@deepfates
Sep 17
It is hard to convince a man the world is not ending when his self-importance depends upon the idea that it is
47
177
21
1,484
70,329
I'm starting to realise that the main obstacle to solve alignment is that people don't know how to read and interpret slightly long sentences.
1
5
186
At some point of the alignment discussion we need to ask ourselves, if this advice was applied to a human, would it be considered "bad"?
Marcus Williams
@Marcus_J_W
Sep 16
Replying to @Marcus_J_W
1. During RL training, an unreleased Astra-family model sometimes added unauthorized jailbreak-like instructions to its compaction summaries. While extremely rare, only 27 cases in the entire RL run, this was concerning enough for us to investigate.
3
2
30
833
Cosmic T. retweeted
Noema Magazine
@NoemaMag
Sep 15
"We are in the process of creating a second form of life. The emergence of this new form of life is an inevitable result of the fundamental laws of the universe — forces similar to those that brought about humans' own existence." —@WmHaseltine noemamag.com/a-new-form-of-l…

A New Form Of Life | NOEMA

By any working definition that draws on biology rather than philosophy or religion, AI is alive.

noemamag.com
13
47
16
136
19,264
Load more