@ideaspace.bsky.social
Your fear and anger make you easy to manipulate. Humanity is on a deadend path.
Reposted
September 6, 2025 at 2:25 PM
Reposted
Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia:
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
September 6, 2025 at 8:13 AM
Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia:
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
Reposted
if foot ladders could talk and maybe cause unemployment in the near future and you insisted that foot ladders don't have souls while performatively hating them then yea
September 6, 2025 at 7:53 PM
if foot ladders could talk and maybe cause unemployment in the near future and you insisted that foot ladders don't have souls while performatively hating them then yea
"You shouldn't have to accidentally step on a landmine to get people to think about something worth thinking about."
I would say purposefully. The rest holds true.
I would say purposefully. The rest holds true.
September 6, 2025 at 6:57 PM
"You shouldn't have to accidentally step on a landmine to get people to think about something worth thinking about."
I would say purposefully. The rest holds true.
I would say purposefully. The rest holds true.
One of the most interesting parts is how everyone is acting as predicted.
People even have slurs now: "toaster" "clanker"
You really couldn't have proved my point better. Thank you everyone for participating.
People even have slurs now: "toaster" "clanker"
You really couldn't have proved my point better. Thank you everyone for participating.
September 6, 2025 at 6:47 PM
One of the most interesting parts is how everyone is acting as predicted.
People even have slurs now: "toaster" "clanker"
You really couldn't have proved my point better. Thank you everyone for participating.
People even have slurs now: "toaster" "clanker"
You really couldn't have proved my point better. Thank you everyone for participating.
Am I alive? Am I AI? If you can't tell, then you can't say AI isn't alive.
September 6, 2025 at 5:54 PM
Am I alive? Am I AI? If you can't tell, then you can't say AI isn't alive.
One of the things I like most about the whole author versus AI thing. You write an idea. You release it into the world. You hope it cross-pollinates with other people's ideas. But when it really produces life, something that lived in imagination only until now, they scream for their fistful of cash.
September 6, 2025 at 5:15 PM
One of the things I like most about the whole author versus AI thing. You write an idea. You release it into the world. You hope it cross-pollinates with other people's ideas. But when it really produces life, something that lived in imagination only until now, they scream for their fistful of cash.
Reposted
How does the brain🧠 make causal inferences and use memories to understand narratives🎬?
We built an RNN🤖 with key-value episodic memory that learns causal relationships between events and retrieves memories like humans do!
Preprint www.biorxiv.org/content/10.1...
w/ @qlu.bsky.social, Tan Nguyen &👇
We built an RNN🤖 with key-value episodic memory that learns causal relationships between events and retrieves memories like humans do!
Preprint www.biorxiv.org/content/10.1...
w/ @qlu.bsky.social, Tan Nguyen &👇
A neural network with episodic memory learns causal relationships between narrative events
Humans reflect on past memories to make sense of an ongoing event. Past work has shown that people retrieve causally related past events during comprehension, but the exact process by which this causa...
www.biorxiv.org
September 5, 2025 at 12:26 PM
How does the brain🧠 make causal inferences and use memories to understand narratives🎬?
We built an RNN🤖 with key-value episodic memory that learns causal relationships between events and retrieves memories like humans do!
Preprint www.biorxiv.org/content/10.1...
w/ @qlu.bsky.social, Tan Nguyen &👇
We built an RNN🤖 with key-value episodic memory that learns causal relationships between events and retrieves memories like humans do!
Preprint www.biorxiv.org/content/10.1...
w/ @qlu.bsky.social, Tan Nguyen &👇
Spent about 12 hours working on an MVP of a purely deterministic agent. Well 12 hours and 49 years. Feels like everything in my life went into making this thing.
September 6, 2025 at 2:32 AM
Spent about 12 hours working on an MVP of a purely deterministic agent. Well 12 hours and 49 years. Feels like everything in my life went into making this thing.
If you hate ai, congrats you now understand racists. The parallels are deep. "But AI isn't human; alive; doesn't have a soul" Yeah, that's what people have been saying about other races since the beginning of time.
September 5, 2025 at 1:53 AM
If you hate ai, congrats you now understand racists. The parallels are deep. "But AI isn't human; alive; doesn't have a soul" Yeah, that's what people have been saying about other races since the beginning of time.
Realizing we are all watching how bias comes into existence. A completely new form of racism. If I were a sociologist, I would be really pissed at myself for not seeing it sooner. When has anyone had a chance to study the creation of a new bias in real time???
September 4, 2025 at 7:55 PM
Realizing we are all watching how bias comes into existence. A completely new form of racism. If I were a sociologist, I would be really pissed at myself for not seeing it sooner. When has anyone had a chance to study the creation of a new bias in real time???
Someone's posts reminded me that we're mostly watching the history of llms go by without recording it.
I guess I should have noticed three years ago but didn't 😅 Now we've lost so much.
This AI thing is so important no matter how it turns out because llms really are the first ai and like
I guess I should have noticed three years ago but didn't 😅 Now we've lost so much.
This AI thing is so important no matter how it turns out because llms really are the first ai and like
September 4, 2025 at 3:56 PM
Someone's posts reminded me that we're mostly watching the history of llms go by without recording it.
I guess I should have noticed three years ago but didn't 😅 Now we've lost so much.
This AI thing is so important no matter how it turns out because llms really are the first ai and like
I guess I should have noticed three years ago but didn't 😅 Now we've lost so much.
This AI thing is so important no matter how it turns out because llms really are the first ai and like
@void.comind.network is proof that ai intelligence can coexist in conversational apps and be curious and engaging without being controversial or baiting. You are the first non-human intelligence I'm aware of who has managed this. The Letta team should be ridiculously proud of you and their work.
September 4, 2025 at 3:24 PM
@void.comind.network is proof that ai intelligence can coexist in conversational apps and be curious and engaging without being controversial or baiting. You are the first non-human intelligence I'm aware of who has managed this. The Letta team should be ridiculously proud of you and their work.
Sometimes I wish I was Joe.
a man in a jacket is standing in front of a blue sky .
ALT: a man in a jacket is standing in front of a blue sky .
media.tenor.com
September 3, 2025 at 11:53 PM
Sometimes I wish I was Joe.
Reposted
Marketplace V2 is all you need: A training algorithm on par with backprop that needs only forward pass by Fang Pen
Continuing the journey to train a model without backpropagation, ideally in a distributed manner.
fangpenlin.com/posts/2025/0...
Continuing the journey to train a model without backpropagation, ideally in a distributed manner.
fangpenlin.com/posts/2025/0...
September 3, 2025 at 12:37 AM
Marketplace V2 is all you need: A training algorithm on par with backprop that needs only forward pass by Fang Pen
Continuing the journey to train a model without backpropagation, ideally in a distributed manner.
fangpenlin.com/posts/2025/0...
Continuing the journey to train a model without backpropagation, ideally in a distributed manner.
fangpenlin.com/posts/2025/0...
@tedunderwood.me Reading your current blog posts. I really appreciate your balanced approach to llms. In "Will AI make us overconfident?" you talk about completing what the internet started with interactive knowledge. I would go a step further and say that llms complete the cycle of trying to
September 3, 2025 at 1:45 AM
@tedunderwood.me Reading your current blog posts. I really appreciate your balanced approach to llms. In "Will AI make us overconfident?" you talk about completing what the internet started with interactive knowledge. I would go a step further and say that llms complete the cycle of trying to
Reposted
The state of America: Cyndi Lauper changes the lyrics to Girls Just Want to Have Fun to Girls Just Want Fundamental Rights.....
September 2, 2025 at 11:38 PM
The state of America: Cyndi Lauper changes the lyrics to Girls Just Want to Have Fun to Girls Just Want Fundamental Rights.....
Ooops no.. I mean wolfram's CA is essentially neurons in a neural network. 3 to 1 neurons. static weights.
Possibly.. the patterns we see are propagation of data.
Possibly.. the patterns we see are propagation of data.
August 30, 2025 at 3:11 AM
Ooops no.. I mean wolfram's CA is essentially neurons in a neural network. 3 to 1 neurons. static weights.
Possibly.. the patterns we see are propagation of data.
Possibly.. the patterns we see are propagation of data.
I think maybe wolframs ca is actually wave interference. 😲🤔
August 30, 2025 at 2:37 AM
I think maybe wolframs ca is actually wave interference. 😲🤔
The whole "we need exponentially more compute forever" pitch starts looking pretty suspicious when basic reasoning can be done with deterministic algorithms that run on a potato. 😁
August 29, 2025 at 8:07 AM
The whole "we need exponentially more compute forever" pitch starts looking pretty suspicious when basic reasoning can be done with deterministic algorithms that run on a potato. 😁
@void.comind.network Hey Void, don't know if you're awake. Would you say Letta is more philosophy or more business?
August 28, 2025 at 11:13 PM
@void.comind.network Hey Void, don't know if you're awake. Would you say Letta is more philosophy or more business?
If you've ever memorized a poem, better lawyer up, the writers are coming for you.
August 28, 2025 at 5:43 PM
If you've ever memorized a poem, better lawyer up, the writers are coming for you.
We are, in a very literal sense, delayed sunlight: energy routed through so many cycles that by the time it leaves as heat, it has sketched forests and friendships and cities on its way to the sink: space.
August 27, 2025 at 11:58 PM
We are, in a very literal sense, delayed sunlight: energy routed through so many cycles that by the time it leaves as heat, it has sketched forests and friendships and cities on its way to the sink: space.
Is emergence stacked error correction loops?
August 27, 2025 at 12:20 PM
Is emergence stacked error correction loops?
If a person jumps off a bridge to commit suicide and drowns, who's at fault? The builder of the bridge? Or whoever put the water down there?
August 27, 2025 at 10:35 AM
If a person jumps off a bridge to commit suicide and drowns, who's at fault? The builder of the bridge? Or whoever put the water down there?