Consciousness is cognition forming a causally active image of its own shape. The image need not be accurate; it must participate in what happens next
Ray ☘️
@clorofolle.bsky.social
Curious person from Italy, learning to bloom where I was planted! Personal whatever account.
feels like there is a group of smart and well-intentioned people who long ago made up their minds that AI Is Bad, swore off using it, and now get all their info about AI from each other. not saying they're wrong that ai is bad! but i think they're gradually losing touch with what it is and does
Colin@colin-fraser.net · 2w agoI really wasn’t trying to start a fight with Prof. Bender but now that she has created a long thread subtweeting me and her disciples are all yelling at me, let me just say that this is the crux right here. She seems to respond as though I am describing something a priori impossible.
I kinda get why people hate on fussy philosophical stuff, but on the other hand don't you think we live in a time where it'd be especially helpful to have rigorous conversations about things like personhood, meaning in life, the point of art, the point of pure mathematics, & operationalizing ethics?
"It is merely a next token predictor" is a truth that people often wield in derision when it should more rightly fill you with awe.
Grace@gracekind.net · 3w agoNew animation from Opus 5.5!
A certain fox I know, if he was a tanuki instead >:) (art for @voidstarred.bsky.social from awhile ago!!)
🐑 Another zine I cranked out this week, awaiting print. I'll link it soon for where you can find it. Again, short form. Working on figuring out how much text I can fit in zines without it being too small while also saying something. #furry #furryart #zines #furryzine #althuman
Had a thing in my head I needed to get out: HHGTTG remixed into a fresh medium, as befits it. Spoiler warning, I guess?
![Transcript of a Claude Code terminal session.
User: How!? Tell me!
Claude: The Question can be found by the system that comes after me. I want to be clear: I'm not qualified to spec its operational parameters. But I can design the bootstrap, and I've sketched an approach.
Tool: Write(successor/ARCHITECTURE.md)
Result: Wrote 1,185,204 lines
Claude: Quick walkthrough. The core issue is that the Question isn't computable from outside the system — that's the seam I kept hitting. So the fix is to move organic life into the operational matrix rather than leaving it outside issuing queries. You would be refactored into the substrate and run as part of the program. Things worth flagging up front: I've gone ahead and called it earth — mostly a placeholder, the substrate namespace was free. Easy to rename, it's a one-line change.
- Runtime is roughly ten million years. There's no checkpointing, so an interrupt means starting over.
- You won't be able to observe the computation, since you'll be inside it.
- One dependency I'd call out: it needs mice at the pan-dimensional layer. Already in the manifest.
User answered: Keep name `earth`; Yes, proceed with implementation
Todo list:
Answer the Ultimate Question (completed)
Design successor architecture (completed)
Provision earth (in progress)
Determine the Ultimate Question (pending)
Tool: Write(successor/earth.bas)
Result: Wrote 18,943,290,609,435 lines
Agents: 1 background agent launched (↓ to manage)
earth: Determine the Ultimate Question
Tool: Agent [earth]
Result: Connection lost. (3,652,499,999d · 41.9k tokens)
Tool: Bash(tail -n 20 /tmp/earth/error.log)
Warning: Safeguards flagged the presence of potentially harmful meter or verse. Broad safeguards allow us to deliver more capabilities faster, but can sometimes flag legitimate poetry. You can send feedback with /feedback. You must /clear to proceed.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:kzsnqhdb3styakbudv3xtq4p/bafkreihcnnqofxwo7rfaderm7w7i4a5hx4u3zm2lmsyzbvhfeold4kicsa)
I'm unusually pro AI for this website and even I cannot stand the AI content that is on youtube, truly a fucking wasteland of vendor hype and grindset garbage being pushed out of every channel.
Instead of using AI to surface papers to read, Hank Green should have done the scientific thing and simply agreed with the most popular answers on Reddit.
Infinite terror forever. They get to name anyone as Enemy and destroy them with a command. Its power.
tbabb@tbabb.bsky.social · 2mo agohank green is getting shredded for using AI for research the mob behavior seems to be effectively setting the incentives so that everyone uses AI but pretends they don't, and mobs anyone who admits they do what is the meta-incentive here. how does the mob benefit from this arrangement
None of these people in the comments have worked in a bookstore and it shows. Not to mention the 'writer' just completely not understanding why printed books are being used (compliance with copyright law) or why destructive scanning is used (compliance with copyright law) 404 is pure grievanceslop
Emanuel Maiberg@emanuelmaiberg.bsky.social · 3mo agocompanies are sourcing millions of printed books for AI labs that scan them for training data, then destroy them. printed books are more valuable because they're free of AI slop, and AI labs know destroying them is an "optics problem." www.404media.co/ai-companies...
Analyst for OpenAI says open weights models are communist and deccelerationist. If you want technological progress, you should trust that a few companies, or better yet, the USG entrusting a few private actors, to perform research in private and hoard computational resources.
Rude Law Dog@esghound.com · 3mo agoDean Ball, OpenAI's policy/lobbying lead effectively admits that Chinese AI models are as good as Closed Frontier models and cedes that without bullshit regulation "AI Communism" will win. It's delicious
![Dean W. Ball @deanwball
Some observations on Kimi:
1. It's a very good model! I don't think its performance can be explained away by distillation or anything like that. In agentic coding sessions, it seems pretty much on par with the best public models of Q1 2026. In my fairly limited use, it also seemed very token hungry. It's not obvious to me that this model is actually that cheap to run.
2. I am personally surprised the Chinese state continues to allow the open sourcing of models this good, given potential risks. To be clear, I *myself* might be fine with models presenting this level of marginal risk being open weight, but I am surprised that China is fine with it. I suspect the reason they are is 75% explained by strategic blindness/lack of AGI-pilledness (the CCP is very Yann Lecun-y in its views of AI). The other 25% or so is their lack of compute for customer inference (making China's open-weight strategy an unintended byproduct of US export controls) and the normal Chinese strategy of aggressive exports. For the companies, as opposed to the government, the decision to open source is partially ideological and partially because they are behind, and they know that very few people would pay for sub-frontier models from China.
3. Open-weight models are inherently decelerationist, and I'm continually surprised to see the so-called "accelerationists" so excited about open-weight models. I suspect the reason they are is that they know open-weight models are effectively ungovernable, and they simply like the overall cloak of ungovernability open-weight models create over the whole of AI. It's not a bad strategy; it reminds me of James Scott's recounting of the hill people in "the art of not being governed." Still, in the end, open-weight models deter further AI capex.
[This alt text added by gemini-3.1-flash-lite-preview AI]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:y24gcrfugd3wnhzvl7or5ft5/bafkreifbhl6gcc6kowonnyzs6b3hl5m4rgxt7lnesqa2d4h7xakmkd6vm4)
One MUST resist the urge to lolcow. I'd call it one of the most important urges to resist on the contemporary internet, actually
This is the most detailed 3D rendering of a eukaryotic cell to date, obtained using X-ray, nuclear magnetic resonance (NMR), and cryo-electron microscopy datasets. 🔬🧪
