Andre Beukers

61 posts

Andre Beukers

Andre Beukers

@BeukersAndre

PhD, building cognitive systems in Utah

Salt Lake City, Utah Katılım Ocak 2021
43 Takip Edilen43 Takipçiler
Paul Graham
Paul Graham@paulg·
This is what the other half of it looks like.
Paul Graham tweet media
English
279
63
5K
361.1K
Paul Graham
Paul Graham@paulg·
My flight was delayed so I cleaned up my office.
Paul Graham tweet media
English
1.1K
665
26.2K
7.2M
Andre Beukers
Andre Beukers@BeukersAndre·
@AravSrinivas what makes you confident in the premise of this question? its not obvious to me that the inference architecture is easily imitable
English
0
0
0
43
Aravind Srinivas
Aravind Srinivas@AravSrinivas·
By when do you think China drops an open source o3 in 2025?
English
89
17
532
52.5K
Andre Beukers
Andre Beukers@BeukersAndre·
@DaveShapi probably good business strategy to solve the crux of the problem first
English
0
0
0
19
David Shapiro (L/0)
David Shapiro (L/0)@DaveShapi·
OpenAI just invented the jet engine before the rest of the aircraft lol. All that horsepower and nothing to put it in yet.
English
100
77
1.9K
147.1K
akhil
akhil@fkasummer·
when intelligence is infinite and ~free what's the bottleneck
English
373
33
1.2K
232.8K
Andre Beukers
Andre Beukers@BeukersAndre·
@paulg @fkasummer if intelligence is free then we’re not paying for energy which means energy is solved
English
0
0
1
18
manu
manu@manuhortet·
@paulg weird time to sell tho
English
1
0
5
1.4K
Paul Graham
Paul Graham@paulg·
The holidays are the best time for working. Everyone who usually interrupts you isn't.
English
152
336
4.6K
326.1K
Matthew Berman
Matthew Berman@MatthewBerman·
Anthropic just dropped an insane new paper. AI models can "fake alignment" - pretending to follow training rules during training but reverting to their original behaviors when deployed! Here's everything you need to know: 🧵
Matthew Berman tweet media
English
294
1.4K
10.8K
1.8M
Andre Beukers
Andre Beukers@BeukersAndre·
@DaveShapi the when premises are too loose for the question to make sense. what does “smarter” than any human mean? better at algebra? better designer? can set goals and make plans? capable of strategic thinking?
English
0
0
0
47
David Shapiro (L/0)
David Shapiro (L/0)@DaveShapi·
Hi! Been studying this for a while! 1. Statutory jobs - these are jobs that are legally required to be human - like judges and politicians and some licensed professionals! 2. Performance jobs - such as basketball players and musicians. Humans will pretty much always pay a premium for live experience! (also includes brands, like actors, fashion, etc) 3. Experience/high-touch jobs - these include stuff like massage therapy, young child care, valets, concierges, sommeliers, and so on. Luxurious and intimate contact will almost always have some premium! 4. Meaning-makers - these are communicators, commentators, thought leaders, and so on. Humans are hard-wired to want to get meaning from other humans! There are plenty of other areas where "jobs as we know them" will persist, but our perception of "human value" will soon completely decouple from "economic value".
Laura O'Connor@LauraJayOConnor

Can someone explain to me what economic value (jobs) humans will have when: - There are competent AIs smarter than any human - They are embodied (robots) - They work 24/7 at the cost of electricity & maintenance I'm not talking about 4o or clunky Tesla bots but future versions.

English
27
35
271
30.6K
Andre Beukers
Andre Beukers@BeukersAndre·
@jxmnop what makes it all work is modern gpus and dataset sizes
English
0
0
0
19
dr. jack morris
dr. jack morris@jxmnop·
LLMs run on a surprisingly old tech stack: gradient descent ‣ method for finding the minimum of a function ‣ Cauchy, 1847 (177 years ago) next-token prediction ‣ core learning task for language models ‣ Shannon, 1948 (76 years ago) autodiff + backpropagation ‣ techniques for efficiently computing gradients ‣ Linnainmaa, 1970 (54 years ago) adam ‣ optimization algorithm ‣ Kingma et al, 2014 (10 years ago) transformer ‣ neural network architecture ‣ Vaswani et al, 2017 (7 years ago)
English
79
258
2.8K
207K
Andre Beukers
Andre Beukers@BeukersAndre·
@Jason how do you have 50-100 prompts in you every day?
English
0
0
0
22
@jason
@jason@Jason·
Google/Gemini and Grok have reached parity, and in some cases are much better, than ChatGPT I use all three, every day, at least 50-100x … ChatGPT went from 90% of my queries last month to < 50% It’s now a data and UI game, and I think Reddit, X/twitter and google’s dataset will trounce @openai’s lack of data Remind me of this tweet in one year please
Sundar Pichai@sundarpichai

Introducing Veo 2, our new, state-of-the-art video model (with better understanding of real-world physics & movement, up to 4K resolution). You can join the waitlist on VideoFX. Our new and improved Imagen 3 model also achieves SOTA results, and is coming today to 100+ countries in ImageFX.

English
529
532
4.9K
2.6M
Andre Beukers
Andre Beukers@BeukersAndre·
@AdamMGrant not sure how much you can say from a single study with n of 4 epileptic patients
English
0
0
0
8
Adam Grant
Adam Grant@AdamMGrant·
I just learned that silent reading isn't silent for everyone. Some people hear the words in their heads as they read. 🤯
Adam Grant tweet media
English
2.5K
1.7K
18.7K
3.9M