mrbungie

47 ms·

271.
by mrbungie 2y ago
People act like Jevons Paradox is an universal law thanks to Satya's tweet.
272.
by mrbungie 2y ago
Not even Microsoft Copilot 365 had a successful launch. The same happened with Apple Intelligence. People talk like the end user demand part of the equation is really solved when invoking an Econ 101 magical interpretation of the Law of Sup
273.
by mrbungie 2y ago
Inference can be more easily offloaded to other kinds of processing units, which also probably are more efficient, like TPUs. That makes both NVDA stock and big AI infrastructure spending less compelling, as those needs are scaled down via
274.
by mrbungie 2y ago
When put into context one could say the OG overreaction was thinking 500B USD in infrastructure investments and restarting nuclear plants were rational decisions rather than working on efficiency.
275.
by mrbungie 2y ago
o1 does not output the full CoT tokens, they are not comparable.
276.
by mrbungie 2y ago
It is still funny. Even if eventually we get there, Jevons Paradox is an observation, not a predictable market strategy. I wouldn't want to depend on it for making decisions. Long term bullish as always, but tech leaders are behaving i
277.
by mrbungie 2y ago
As I see it DeepSeek that totally dirupted USA tech industry PxQ projections and, the price part at least, exposing the fact that more efficient cheaper models are possible. They're coping saying that at a lesser P, inevitably there wi
278.
by mrbungie 2y ago
I'm not sure if it has to do with the UI redesign and renaming, but I don't feel the new app is as reliable as before. My RDP connections now fail all the time.
279.
by mrbungie 2y ago
Maybe DeepSeek ain't it, but I expect a big "box of scraps"[1] moment soon. Constraint is mother of invention, and they are evading constraints with a promise of never-ending scale. [1] https://youtu.be/9foB2z
280.
by mrbungie 2y ago
Metacommentary: I stopped arguing because this thread is reeking of AI bro non-sense, they can't argue with facts so they continue with vague statements and deflection (e.g like the "Whatever" you've got above). The GenA
281.
by mrbungie 2y ago
I'm 85% confident that a paperclip maximizer scenario will include Clippy in a way or another.
282.
by mrbungie 2y ago
1. For me transformer multimodal processing is still language, as the work is done via tokens/patches. In fact, that makes its image processing capabilities limited in some scenarios when compared to other techniques. 2. When on a bike
283.
by mrbungie 2y ago
The question/argument was well defined by parent-grandparent posts: people reason as LLMs, including a suggestion that people discussing here are not being self-conscious of said similarity (i.e. people parroting/regurgitating). Y
284.
by mrbungie 2y ago
This post is full of speculation. For starters, human minds do more than just language. There are some arguments (Moravec's paradox) that posit that language is easy, while keeping the rest of the body working is the real hard problem.
285.
by mrbungie 2y ago
It isn't that simple, the substrate is totally different. The brain is not a GPU, not by any stretch, even if it "runs something analogue to a LLM" inside.
286.
by mrbungie 2y ago
You can't win this argument through logic less mere rhetoric. You need scientific evidence. Hint: There isn't. Edit: Ok, let me be less harsh and play with your argument. Can an LLM have an acid/psilocybin/ketamine trip,
287.
by mrbungie 2y ago
Neuroscience and evidence be damned, [the brain is a computer] is being transformed into [the brain is an LLM]. Happens with every new technology. Edit: What is happening with people these days? They seem to be reducing people and their min
288.
by mrbungie 2y ago
Bit/media rot exists, and it is a matter of time until we lose those bits (or however the signals are encoded) to entropy. Considering the fact that many pieces of media in streaming services disappear all the time, I think we're
289.
by mrbungie 2y ago
Hyperscalers need to justify their current GPU investments with pay2go and provisioned throughput LLM usage revenue. If models get more efficient too quickly and therefore GPUs less loaded by end users, short of a strong example of Jevon&#
290.
by mrbungie 2y ago
I know and I think the same as you, I'm just reproducing what other posters are arguing ITT.
291.
by mrbungie 2y ago
I only know about Moore Threads GPUs. Last time I took a look at their consumer offerings (e.g. MTT S80 - S90), they were at GTX1650-1660 or around the latest AMD APU performance levels.
292.
by mrbungie 2y ago
Chinese companies smuggling embargo'ed/controlled GPUs and using OpenAI outputs violating their ToS is considered cheating. As I see it, this criticism comes from a fear of USA losing its first mover advantage as a nation. PS: I&#
293.
by mrbungie 2y ago
And (some people here are saying that)* if they are up-to-date is because they're cheating. The copium itt is astounding.
294.
by mrbungie 2y ago
That is not going to happen without currently embargo'ed litography tech. They'd be already making more powerful GPUs if they could right now.
295.
by mrbungie 2y ago
We will have to wait to get some info on that probe. I know SMCI is not the nicest player and there is no doubt GPUs are being smuggled, but that quantity (50k GPUs) would be not that easy to smuggle and sell to a single actor without raisi
296.
by mrbungie 2y ago
Then the question becomes, who sold the GPUs to them? They are supposedly scarse and every player in the field is trying to get ahold as many as they can, before anyone else in fact. Something makes little sense in the accusations here.
297.
by mrbungie 2y ago
This. One of the main challenges in GenAI today is UX: how do you actually embed the "intelligence" of LLMs into software in innovative ways that are not mere chatboxes, because prose sucks for showing information and using said o
298.
by mrbungie 2y ago
Getting from 2% (Oracle) to 10% (GCP) market share would need 37.97% CAGR in 5 years. In a vacuum where everything else keeps the same, maybe, but I see that goal as very difficult to attain in what is a highly competitive industry right no
299.
by mrbungie 2y ago
Oracle has 2-3% market share as a Cloud Provider. MSFT or even Google (AWS is not as mature in that space imho) made perfect sense, Oracle doesn't. Elon and Larry are good friends, I would guess that has something to do with this devel
300.
by mrbungie 2y ago
Softbank is not exactly a green flag when using their involvent as a signal of "low hypeness". I still remember WeWork.

More ›