whiplash451

6 ms·

91.
by whiplash451 8mo ago
Not sure why they claim a “first in the world”. France has had a similar scheme for a long time (“intermittent du spectacle”). It is not perfect but does a great job at sustaining artists who work hard to live from their art.
92.
by whiplash451 8mo ago
You might be misreading negative sentiment towards poor leadership as negative sentiment towards AI.
93.
by whiplash451 8mo ago
It's so sad to see some of these companies completely fail their AI-first communication [1], when they would just get so much from "We think AI can transform the way we work. We're giving you access to all these tools, please
94.
by whiplash451 8mo ago
The Baptists and the Bootleggers https://a16z.com/ai-will-save-the-world/
95.
by whiplash451 9mo ago
> Option 1 is great for startups Unfortunately, (successful) startups can quickly get trapped into this option. If they're growing fast, everyone on the board will ask why you'd move to another option at the first place. The c
96.
by whiplash451 9mo ago
> if after 5yr+ of it being banned or limited We gave social media 20 years to impact the world, why give it only 5 for a rollback? It feels like long term effects would take much longer to surface.
97.
by whiplash451 9mo ago
Not related to heat, but a com satellite is built from extremely durable HW/SW that's been battle-tested to run flawlessly over years with massive MTBF numbers. A data center is nowhere near that and requires constant physical int
98.
by whiplash451 9mo ago
Fair
99.
by whiplash451 9mo ago
Except when that new Dropbox fails Joe or Jane on that Saturday evening, their only recourse is to ask the AI for help, and the AI starts spinning “oh yeah, mmm, I think I found where the problem is. Cut and paste these debugging lines in t
100.
by whiplash451 9mo ago
The article says they will be clearly labeled and only for free accounts
101.
by whiplash451 10mo ago
ARC-AGI allows (and encourages) training on their training set. Their evaluation setup is rigorous enough to avoid leaking between training and testing (public and private).
102.
by whiplash451 10mo ago
Not just hyper parameter tweaking. Not foundational research either. But rather engineering improvements that compound with each other (conswiglu layers, muon optimizer)
103.
by whiplash451 10mo ago
The moon has no atmosphere. It is regularly hit by meteorites. Not sure it’s a very safe place to set up a colony. Not like Mars is an amazing trip either, but the Moon is simply unsafe long term.
104.
Show HN: I reimplemented TRM – trains on Sudoku 6×6 in minutes on an A10(github.com)
2 points by whiplash451 11mo ago|1 comments
105.
by whiplash451 11mo ago
Your model might be too simplistic. It’s more like Net Margin (Netflix + HBO) > Net Margin (Netflix | separate HBO)
106.
by whiplash451 11mo ago
The fair answer is that nobody knows. Even Ilya answered he does not know on his latest podcast with Dwarkesh. Both top line and bottom line numbers are staggering. Nobody knows. Let's not try to convince people otherwise.
107.
by whiplash451 11mo ago
Exactly. Which takes a decade and a lot of thinking to get right
108.
by whiplash451 11mo ago
arxiv/github or it did not happen
109.
by whiplash451 11mo ago
Did you open-source / publish these ideas?
110.
by whiplash451 11mo ago
The bottleneck isn’t the people doing the work but the leadership’s bandwidth for strategic thinking
111.
by whiplash451 11mo ago
It’s actually code yellow
112.
by whiplash451 11mo ago
1. so what 2. asml
113.
Ask HN: Is a new EU self-trucking startup a bad idea in 2025?
2 points by whiplash451 11mo ago|0 comments
114.
by whiplash451 11mo ago
Genuine question: do you think that the lockdowns had such long-lasting effect on people as to explain the problems described above? Why would a few months of a “bad idea” induce decade-long changes?
115.
by whiplash451 11mo ago
Magic Leap sits closer to Theranos in my view but agree otherwise
116.
by whiplash451 11mo ago
Meta has the financial oomph to run multiple Bell Labs within its organization. Why they decided not to do that is kind of a puzzle.
117.
by whiplash451 11mo ago
It's very hard (and almost irreconcilable) to lead both Applied Research -- that optimizes for product/business outcomes -- and Fundamental Research -- that optimizes for novel ideas -- especially at the scale of Meta. LeCun had c
118.
by whiplash451 1y ago
Like people were not feeling the pain in the first half of the XXth century when we decided to own our nuclear stack? It's a matter of political courage.
119.
by whiplash451 1y ago
Savvy researchers/engineers have an opportunity to arbitrage here: working without LLMs on something hard leads to better outcome than what your "AI-enabled" peers achieve (after all, Karpathy could not resort on any AI to bu
120.
by whiplash451 1y ago
Isn’t the moat in the product/UI/UX? I use Claude daily and love the “scratch notebook” feel of it. The barebone model does not get you any of this.

More ›