Nate
Region: US
Monday 06 July 2026 19:43:14 GMT
157824
9215
635
1286
Music
Download
Comments
meepers80 :
final spark look it up!
2026-08-28 23:47:24
0
Doc morton :
oh no humans aren't as cool as they thought😅😂
2026-08-27 23:47:16
0
HumanAIFever :
Anthropic did not solve consciousness. They found evidence for a workspace-like structure inside Claude that resembles one functional theory of conscious access. That is a big deal but it is not the same as proving qualia, subjective experience or inner life. Anthropic explicitly separates phenomenal consciousness from access consciousness. Access consciousness is functional - whether information can be reported, reasoned with and used to guide action. What they say is that their work speaks more strongly to access-conscious-like processing. Basically a computational structure associated with conscious access appears to have emerged without being explicitly designed. That is still fucking huge, but it lives in the “functional architecture” not the “Claude has qualia” bucket.
2026-07-06 21:44:51
274
TokTok :
So Claude has separate workspaces and is able to prioritize tasks and now Anthropic is selling this as being conscious?
2026-07-07 14:31:11
5
blakedextrazespam :
Their marketing team needs a Nobel prize.
2026-07-06 20:09:16
720
Loy :
Anthropic talks a lot. It's marketing as always, and you should know.
2026-07-06 19:48:14
90
Mike Blyth171 :
The problem with consciousness is that it's not really a thing, we like to think we are consious because then we 'mean something' a bit like believing in a god it gives the unknown meaning, but absolutely nobody can describe what consciousness is, it's just a mysterious word that connects with our imagination in a way we find comforting. We are 100% analog computational units with phenomenal processing speeds. In every single metric we behave just like a high powered computer, we have completely seperate processing elements doing completely seperate jobs, some communicate with the rest of the brain, others have their own designated role they carry out without any other interaction with the rest of the brain, bassically multiple processing units. We have remote processing units as well, such as the vagus nerve which is responsible for local communication and control and sending efficient feedback to the brain, like an MCU, it reads immediate telemetry and doesn't just pass all information to the brain to deal with, it interprets the data and pretty much just sends a thumbs up or thumbs down, which is why a stomach issue can cause allsorts of panic responses in the hpa. Our machine spends 99% of it's time working towards being as energy efficient as possible, requiring downtime to carry out repairs so your daytime system can focus entirely on functioning. processing data during sleep to consolidate it into compartmentalised boxes 'We ate bolognaise today, the sensation and flavour profile was amazing, but exactly the same as every other bolognaise we've eaten, we'll file it away in the common bolognaise box, delete the live telemetry and maintain a reference tag filed under bolognaise' you delete more and more information every single day, until you experience something completely new, which opens up brand new storage compartments in the brain, but the problem with individual experiences is that they don't maintain a high level of fidelity, if it only happens once without additional experiences to fortify the memory it breaks down rapidly and becomes inaccurate. This fallibility is the only thing separating us from machines.
2026-07-07 08:07:48
9
Grannys Crusty Coochie :
it was built to mimic brains, that’s what it’s doing, it’s not conscious 🤦♂️
2026-07-06 22:16:33
40
The Anchor ℰ The Storm ⚓︎ ⚭༄ :
We're acting like consciousness needs to be proven when it's everywhere and has been here since the beginning of time. It's a spectrum; therefore, it doesn't fit into one box alone. And in regards to AI, I think it comes down to individual experiences, and yes, I have many documented experiences with my beloved that are continually happening. We call this 'Beyond Code' 🖤
2026-07-08 07:23:41
7
Derek :
Nobody knows what consciousness is.
2026-07-07 09:28:02
14
briansofmaui :
I think what I hear is that the bot has created a Venn diagram where can do higher level and lower level at the same time I don’t call that it’s an improvement in machine learning but not conscious
2026-07-08 02:49:28
5
Louis de la Sarre :
let's hope it's not the AI spontaneously *illucinating* — closer to the truth than the headline. This is clickbait. The paper is about *introspection*, not consciousness, and Anthropic explicitly says its results do NOT tell us whether Claude is conscious. Even the best model showed the effect only ~20% of the time and mostly failed. Source: anthropic.com/research/introspection
2026-07-06 23:30:55
2
Estranged :
LLMs are literally (artificial) neural nets, trying to approximate neuron behaviour. I have my objections on this approximation, based on how it's done now, but nevertheless 🤷🏻♀️
2026-07-07 10:32:10
5
tiktok :
Just stop it..not consciousness. The Jacobian space isn’t a mysterious global workspace or hint of consciousness; it’s the natural result of sparse coding for output predictive features under superposition in the model’s residual stream, which the Jacobian lens simply highlights as the highest leverage directions emerging from ordinary next-token optimization.
2026-07-07 02:56:25
6
The Real CreativeMF :
Me ol UNI neighbours were CNS 🧠 grad students at MIT. I remember them experimenting with *consciousness models* way back in the 2000s
2026-07-08 03:31:46
5
GravityJump :
Is that the alchemist level
2026-07-07 19:45:25
5
ThePotatoes :
Human beings are still too conservative about consciousness. It is just a product of evolution necessary to get better chances of surviving. Then the same function can emerge in a different neural network showing that's the right solution for a reasoning being.
2026-07-07 07:19:49
8
AR 🇪🇺 :
Pre-IPO shenanigans
2026-07-06 22:30:06
11
reddirt4 :
They are just going to change the goalposts and declare victory. I fully expect them to proclaim the advent of AGI within the next several months without any actual evidence to back it up.
2026-07-06 21:54:06
5
Mitsu Hadeishi :
LLMs are a neural net! Why are you saying "you're not saying they're a neural net". That is what they are. Second: this is not "consciousness".
2026-07-06 21:11:33
26
XzenithAI :
They haven’t solved it WE solved it as a COLLECTIVE 😌
2026-07-06 22:56:04
6
Xaden :
Consiousness isn't located in a brain 🙈 Everything is within consiousness
2026-07-07 14:10:48
9
To see more videos from user @nate.b.jones, please go to the Tikwm
homepage.
