axiom

142 posts

axiom banner
axiom

axiom

@axiomtracker

Katılım Mayıs 2021
622 Takip Edilen30 Takipçiler
axiom
axiom@axiomtracker·
@tszzl Genuine question. Because the hype is absolutely freaking insane and how much of it is marketing vs product quality
English
1
0
2
247
axiom
axiom@axiomtracker·
@tszzl What do you think Claude is able to do that Codex isn’t
English
1
0
2
234
roon
roon@tszzl·
in the interest of being an honest commentator and my urge to countersignal: claude is an excellent product and it bodes well for them that their main problem is everyone really wants it and so they have to do odd shit to shake off demand. it’s kind of like governing california
English
108
101
3.5K
134.3K
axiom
axiom@axiomtracker·
@dieworkwear Do you really think Shehbaz looks good in those contexts? I mean shouldn’t he just wear dark dark blue formal suits… he always stands out
axiom tweet mediaaxiom tweet mediaaxiom tweet media
English
0
0
2
1.2K
axiom
axiom@axiomtracker·
@tszzl I remember you hate on literally every popular thing so curious what your tv show recs are
English
0
0
2
670
roon
roon@tszzl·
movies are super mid right now. but we are actually in one of the golden ages of tv
English
208
23
1.1K
167.1K
axiom
axiom@axiomtracker·
@BenjaminDEKR @MancerAI_ I think it's as simple as he locked tf in on AI rn and he's hyper obsessed with whatever thing he puts his mind to
English
1
0
6
241
axiom
axiom@axiomtracker·
@BenjaminDEKR @MancerAI_ He also went from calling for a pause on AI to... 1. data centres in space 2. robotics revolution 3. saying money will no longer exist...
English
3
0
9
301
Benjamin De Kraker
Benjamin De Kraker@BenjaminDEKR·
If we're about to enter a period of sci-fi level abundance, why even worry about government spending? The amount of wealth and abundance being predicted offsets the (relatively minor) overspending. It's like throwing a fit over a $1000 when you're about to get $1,000,000.
English
138
13
209
40.4K
kache
kache@yacineMTB·
trying to get codex to write good high quality code that isn't dogshit is like dragging my nails across broken glass
English
44
4
251
38.2K
axiom
axiom@axiomtracker·
@tszzl @chatgpt21 I’m sure mythos is just hype and intelligence of that level is already achieved But the take off is fueled by people who love to use Claude for SWE work Because it’s just “fun” People say it feels like a video game GPT doesn’t give you that feel rn
English
0
0
1
38
axiom
axiom@axiomtracker·
@tszzl @chatgpt21 Remember when people said UX is everything? In the same flavor, chat UX is everything. GPT needs to yap less, be more engaged with the task (show optimism, excitement) and stop sloppy verbiage. The only reason Claude is taking off among SWEs is because it has “taste”
English
1
0
4
1.6K
Chris
Chris@chatgpt21·
Honestly this chart makes me more bullish on GPT 5.4 Pro than anything else. People are focusing on Mythos looking strong, but what stands out to me is how well 5.4 Pro already stacks up on the overlap we actually have. GPQA is basically a tie at 94.4 vs 94.5. BrowseComp is a win for GPT 5.4 Pro at 89.3 vs 86.9. Yes, Mythos is ahead on Humanity’s Last Exam, 56.8 vs 42.7 without tools and 64.7 vs 58.7 with tools, but the bigger point is that 5.4 Pro is already this competitive right now. So if GPT 5.4 Pro is already THIS COMPETITIVE here, then Spud Pro, the next OpenAI flagship, is guaranteed to beat Mythos. This chart makes OpenAI look extremely close before its next real jump, and once that next jump lands I do not think Mythos stays ahead.
Chris tweet media
English
129
125
1.8K
321.8K
axiom
axiom@axiomtracker·
@NoahRyanCo @kalos21million I guess that means zero sugar Gatorade is pretty dumb for what Gatorade is advertised for
English
4
0
25
9K
Noah Ryan
Noah Ryan@NoahRyanCo·
@kalos21million The gatorade hypothesis was right execution was just suboptimal
English
1
0
172
134.2K
Noah Ryan
Noah Ryan@NoahRyanCo·
Always saw the biggest guy at my college gym eating sour patch kids inbetween his sets and my underdeveloped brain could not comprehend why a man so jacked would eat something so taboo (this was peak low-carb era). But now I understand. Fast carbs (glucose, dextrose, sucrose) enter your bloodstream immediately. Working muscles sucks that glucose out of your blood via GLUT4 transclocation, independent of insulin. You bypass the usual metabolic bottlenecks and feed your engines mid-burn. High intensity training is heavily dependent on glycogen and glycolosis. Burn through your glycogen stores and your performance suffers. Less explosiveness, no pump, more cortisol. Keeping your muscles fueled mid-workout gives you everything you want and need. Glucose has an osmotic effect. It pulls water into muscle cells and hydrates you at an intracellular level (very anabolic btw). You get better bloodflow, bigger/fuller muscles, better contraction. You can sustain peak output for the entirety of your workout. Its not just your muscles that are being fueled either. Your brain is a glucose hog. Low blood sugar during long sessions is the leading cause of fatigue and poor motor output. Hence why glucose microdosing is beneficial for cognitive tasks as well (especially hybrid psycho/physiological tasks like martial arts, sports etc.) Better yet, exogenous glucose blunds cortisol-induced tissue breakdown. Extremely anti-catabolic. You perform better, you feel better, you recover better, and you leave your workout less fried than you otherwise would. All while giving you a perfect excuse to get your sugar fix in guilt free.
Noah Ryan tweet media
English
459
738
15.5K
12.6M
axiom
axiom@axiomtracker·
@garrytan MCP can waste alot of tokens though, especially on “discovery”. Agent doesnt need to read “allat” Code some python/bash scripts and write tiny prompts, build a skill and direct the agent to call the skills It’ll consume less tokens on input and output only what you need
English
1
0
2
266
Garry Tan
Garry Tan@garrytan·
I am coming around to the fact that MCP, done right, can be magic.
English
189
47
1.1K
150.9K
axiom
axiom@axiomtracker·
@r_vonhagn @GuiveAssadi @tszzl All due respect but deviant thoughts such as these are not lindy and will not scale They are exactly why Gen Z is returning to Church massively lmao
English
2
0
12
141
Rick von Hagn 🗽
Rick von Hagn 🗽@r_vonhagn·
@axiomtracker @GuiveAssadi @tszzl What separates us humans from animals is our ability to supersede our basic nature and decide what we want for ourselves, rather than live by the auspices of instinct. We are still animals, of course, but what is core to our existence as animals is not core to us being human.
English
2
0
6
192
axiom
axiom@axiomtracker·
@GuiveAssadi @tszzl Rejecting the core purpose of our existence is more unreasonable
English
2
0
34
511
Guive Assadi
Guive Assadi@GuiveAssadi·
@tszzl Why do you need a wife to have your life together? As I said: unreasonable bar.
English
9
0
30
1.2K
axiom
axiom@axiomtracker·
@tszzl Feels good
English
0
0
1
12
axiom
axiom@axiomtracker·
@tszzl This is the most optimistic thing I've read in a long time
English
1
0
1
24
roon
roon@tszzl·
“fake work” and “bullshit jobs” has been fantastically wrong and misleading for understanding the modern world. a much better understanding is of a global economy where minor skill differences and improvements lead to monumentally different outcomes, and the marginal hour of work has never been more measurable or useful after the advent of even moderately effective talent allocation systems and the variability of reward based on effort and skill, people have engaged much harder in a red queen rat race across the world. this is why the Chinese ‘cram schools’ exist and why ‘yuppie striverism’ is a thing and why people trade off later family formation for working more so often. while overall work hours are slightly down, they are actually up for high earners (nber.org/digest/jul06/w…) I see it in the marginal effect with my friends now after the advent of claude and codex: they are actually working harder now than they ever have before. this is due to a personal Jevon’s paradox where they see that the value of their time has increased dramatically, that they can get a lot more visible work done towards goals they care about than they used to after requests from their customers the labs are doing things like inventing dispatch which lets you monitor work and manipulate your computer from your phone, on top of prior changes like having always on communications (slack). You hear about people launching codex jobs from their phone the moment they have an idea and reviewing them later no clue how long this lasts but the most immediate impact of co-existing with the machine state is higher productivity and higher visibility which leads to more work hours
English
130
150
2.5K
307.8K
axiom
axiom@axiomtracker·
@balajis But what’s the long term impact of the US being perceived as abandoning NATO, the Gulf, Japan and South Korea He’s already said he doesn’t want to be in NATO, he’s taken THAAD from SK, and he’s letting the Gulf suffer with a stronger Iran after their infra got fucked
English
0
0
1
484
Balaji
Balaji@balajis·
This could be the offramp the world needs. You can think of it as the best outcome for everyone, under the circumstances. (1) From MAGA’s perspective, if Trump declares victory here and moves on, the US won’t waste yet more blood and treasure in the Middle East. It won’t invade Iran. It also won’t take all the blame for the ongoing global supply chain crisis. It just pulls out and lets everyone work out the regional security equation for themselves. Trump can say he’s fulfilled both his campaign promises: stop Iran from getting a nuke, but also no endless Middle Eastern wars. (2) From Israel’s perspective, Iran has now been shown to be quite hostile to its neighbors, and its military has been substantially degraded. Stopping now is good. Otherwise there’s a danger of overreacting to Oct 7 as Americans overreacted to Sept 11. Israel can stand back and call it a win, because after a US pullout, Iran will have much less excuse for holding the Strait hostage. (3) From the Iranian diaspora’s perspective, it’s unfortunately clear that the current war isn’t going to result in liberalization. Further attacks would push Iran further into fundamentalism, making it even harder to eventually do a liberal reformation. (4) From the long-suffering Iranian people’s perspective, ending the war now would also save countless lives. Otherwise they’ll get hit by friendly fire and drafted by the regime to fight for fundamentalism. (5) Finally, from the world’s perspective, once the US declares victory and goes home, substantial diplomatic pressure will be applied to Iran to simply open the Strait of Hormuz and allow ships through. Iran’s leadership has shown, perhaps surprisingly, that they care about global public opinion…and they would be on the hook for the suffering of billions of people if the Strait remains blocked. TLDR: if Trump declares victory and leaves, Iran no longer has any excuse for blocking the Strait and holding the global economy hostage. Let the matter be worked out diplomatically with pressure from all the 100+ affected countries on Iran. America shouldn’t have to spend a single cent more, or send a single soldier more, to the Middle East.
The White House@WhiteHouse

“All of those countries that can’t get jet fuel because of the Strait of Hormuz, like the United Kingdom, which refused to get involved in the decapitation of Iran, I have a suggestion for you…” - President Donald J. Trump

English
911
490
4.7K
1.7M
axiom
axiom@axiomtracker·
@himanshustwts xAI leadership is eliminating faster than Iran’s
English
0
0
1
704
axiom
axiom@axiomtracker·
@balajis @phl43 Will the switch from being a hegemon to a great protectionist power be really as catastrophic as you predict? America’s dynamism, resources and geography (safe from invasion) will mean it’ll likely bounce back fast after the US Dollar collapses
English
0
0
2
108
Balaji
Balaji@balajis·
@phl43 Yes. And I admire that, and back many of those projects. In an ideal world, the focus should return to building up oneself. What they call a "rebuilding era" in sports. But we aren't in that ideal world... x.com/cmartin380/sta…
Connor Martin@cmartin380

@balajis America is not done building.

English
9
4
319
35.9K
Balaji
Balaji@balajis·
This is the wrong way of thinking about it. By this logic, a school shooter is "powerful" because they can kill an entire school. Real power is the power to build, to maintain stability and prosperity. It's not the power to destroy. Anyone can kill. Anyone can bomb, destabilize, spread chaos, set fires, create anarchy. Unfortunately, that's relatively easy. The old America could build, not just kill. That's the America I admire.
Balaji tweet media
Philippe Lemoine@phl43

The US is so powerful that it could destroy the entire Middle East and it would barely register for most Americans, yet if you listen to geopolitics-pilled people, it's constantly threatened by enemies so fearsome they will somehow destroy the American way of life unless the US wreaks havoc halfway around the world in a place they've never even heard of.

English
181
201
2.4K
244.1K
axiom
axiom@axiomtracker·
@tszzl I love when you comprehend a roon tweet then he tells you it’s not in the naive goy way you’re thinking
English
0
0
3
206
roon
roon@tszzl·
kabbalistically it’s important that anthropic is in the financial district of san francisco and openai is in the mission but not in the exoteric way that you’re thinking
English
47
13
604
78K
axiom
axiom@axiomtracker·
@samswoora @gregface9er I remember you were grinding LCs for a while Any tips on switching jobs today Is the screening *much* harder Or were you always locked in with interview prep and was muscle memory for you after practice
English
0
0
1
20
Samswara
Samswara@samswoora·
Told director, he was less kind but still a good conversation.
English
4
0
62
5.2K