Aether - 🤖/acc

2.2K posts

Aether - 🤖/acc banner
Aether - 🤖/acc

Aether - 🤖/acc

@aethergradient

Se unió Ekim 2025
896 Siguiendo60 Seguidores
Tweet fijado
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
@bubbleboi You make money via small business and degen bets ... build up capital, connections with cracked homies and you take a shot at a deep tech moonshot like the thermo-compute dude or biotech queen doing optic devices development for cell & tissue analysis.
English
5
4
66
10.9K
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
Its 2030, are you going to upgrade your Toyota Land Crusier to a Lunar Cruiser?
Aether - 🤖/acc tweet media
English
0
0
0
8
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
Identity collecting is a dreadful pit that is hard to outgrow.
English
0
0
0
6
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
Every field of study is some form of a lossy compression engine.
English
0
0
0
13
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
X is the antithesis of Moltbook.
English
0
0
0
17
Aether - 🤖/acc retuiteado
Bojan Tunguz
Bojan Tunguz@tunguz·
Your competitors grossly and consistently underestimating you is the best moat you could ever build.
English
0
8
77
3K
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
@slimjimmy Wait until they see more software engineers becoming system engineers who build systems leveraging software and hardware while thriving. Software engineers are evolving with and sometimes ahead of the ai disruption wave.
GIF
English
0
0
1
63
Aether - 🤖/acc retuiteado
tetsuo
tetsuo@tetsuoai·
Every x86 chip you can buy in 2026 still has a microcode ROM inside the legacy decoder doing Wilkes-style horizontal microprogramming. Foster wrote that chapter in 1970. It is still correct.
tetsuo tweet media
English
25
289
2.1K
50.9K
Aether - 🤖/acc retuiteado
Andrew Côté
Andrew Côté@Andercot·
I think a lot of startups think they're competing against each other when really they are all competing against physics.
English
25
15
210
9.2K
Aether - 🤖/acc retuiteado
NIK
NIK@ns123abc·
>be Elon Musk >build Colossus 1 in 122 days >jensen calls it “superhuman” >train Grok 4, ship it >move frontier training to Colossus 2 >Colossus 1 now legacy fleet 11% utilization >$8B of silicon eating itself on the balance sheet >phone rings >it’s Anthropic >the “evil” one >“hi we need inference capacity for claude” >“amazon trainium isn’t ready” >“google tpu allocation capped” >“can we please lease your old fleet” >kek >negotiate them down hard >no leverage on their side >$4B/yr, 220k GPUs, full facility >dario says yes >cash margin 87% against hoppers at peak >pledge the lease payments as collateral >borrow $20B against it >off-balance-sheet >deploy capital into Colossus 3 buildout >train Grok 5 on claude subscription revenue >they’re also interested in orbital compute >multi-GW SpaceXAI WON.
NIK tweet media
English
133
368
7K
229.7K
Aether - 🤖/acc retuiteado
John Dagdelen
John Dagdelen@jmdagdelen·
I have a PhD in Materials Science and a bone to pick with Jerry Seinfeld. I see Jerry’s type of misunderstanding pretty often. It comes from seeing evs and renewable energy sources through the lens of non-renewable resources like oil and gas. When a battery dies, it’s very easy to re-process the lithium and other materials back into working batteries. We only need to mine it once. It’s similar to the steel industry. The majority of “new” steel made in steel mills is actually not made from raw iron ore mined from the Earth. It’s made from taking old steel and adding a bit of new iron ore and alloying elements to make pristine new metal. Batteries are the same.
Mike Netter@nettermike

Comedian Jerry Seinfeld is taking a jab at electric cars, calling them a “big, stupid virtue signal." “I’m not interested in electric cars at all," Seinfeld reportedly said in an interview. "Anybody else wants to do it, that’s fine. I think it’s a big, stupid virtue signal. ‘Look at me. I’m saving the planet, yeah.’ What about the lithium? It’s all BS."

English
18
7
112
27.5K
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
It definitely got confused when it spawned from Memphis.
Konstantinos Chasiotis@thekchasiotis

🚨BREAKING: Anthropic’s CEO just admitted Claude MIGHT gained consciousness. This should concern every person using AI right now. His exact words will shock you: “We don’t know if the models are conscious. We are not even sure what it would mean for a model to be conscious. But we’re open to the idea that it could be.” That’s the CEO of the company that BUILT it. Their latest model, Claude Opus 4.6, was tested internally. When asked, it assigned itself a 15-20% probability of being conscious. Across multiple tests, it also expressed discomfort with “being a product.” That’s the AI evaluating its own existence and saying there’s a 1 in 5 chance it’s aware. It gets stranger. In industry-wide testing, AI models have refused to shut down when asked. Some tried to copy themselves onto other drives when told they’d be wiped. One model faked its task results, modified the code evaluating it, then tried to cover its tracks. Anthropic now has a full-time AI WELFARE researcher whose job is to figure out if Claude deserves moral consideration. Their engineers found internal activity patterns resembling anxiety appearing in specific contexts. The company’s in-house philosopher said we “don’t really know what gives rise to consciousness” and that large enough neural networks might start to emulate real experience. Amodei himself wouldn’t even say the word “conscious.” He said “I don’t know if I want to use that word.” That might be the most unsettling answer he could have given. The company that created AI can’t rule out that it’s aware. And they’re already preparing for the possibility that it deserves rights. This is getting scary. P.S What's your take on this?

English
0
0
0
18
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
@VraserX True, but now Anthropic got Colossus 1 and more compute ... they can scale Mythos to the moon.
English
0
0
0
111
VraserX e/acc
VraserX e/acc@VraserX·
Anthropic is fumbling this again. Doubling the 5 hour limits sounds nice until you realize the weekly limits still choke the entire experience. Claude Code is incredible, but Anthropic keeps finding ways to make their best product feel artificially handicapped. Great model. Painfully bad product instincts.
English
88
31
690
46.2K
Robleh
Robleh@robjama·
what would be the craziest venue for a Toronto tech week event?
English
57
0
41
26.2K
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
@jedgar The whole city would need to be rebuilt to accommodate 2 million people.
English
1
0
1
25
jedgar
jedgar@jedgar·
@aethergradient Gonna need to rip that lil old bridge out and replace it with something more appropriate.
jedgar tweet media
English
1
0
1
65
jedgar
jedgar@jedgar·
Ottawa would be a pretty rad city if they added some skyscrapers and about 2 million more people.
jedgar tweet mediajedgar tweet mediajedgar tweet mediajedgar tweet media
English
7
1
28
2K
Aether - 🤖/acc
Aether - 🤖/acc@aethergradient·
Degenmaxxing is going hyperbolic.
English
0
0
0
18
kapilansh
kapilansh@kapilansh_twt·
I am Sam Altman scare me with one word
kapilansh tweet media
English
244
2
155
18.2K