sarjann

7 ms·

1.
by sarjann 4d ago
Not sure if I missed something but not sure what justification he gave that Open Weights would improve safety. The counter argument from Anthropic is easy. If we open weights we can't detact it after being published / the guard ra
2.
by sarjann 6d ago
Been using bedrock agent core and seems to work fine for me. Although there might be a better abstraction.
3.
by sarjann 6d ago
I still can't stand the fact that in order to use their mobile app you must allow sharing data for training. No thanks will close the app and it seems like they follow the same thing with their desktop app. Outside of that, if I want t
4.
by sarjann 12d ago
And if they can't give capacity, they should maybe be required to give a kind of "call option" or hedge locals energy prices linked to regular inflation.
5.
by sarjann 13d ago
"It is the training apparatus of your future senior staff, and you should protect it the way you’d protect any other piece of critical infrastructure." I think a problem is that especially with the culture of job hopping a lot and
6.
by sarjann 14d ago
Why use electricity alone for open models? Surely you’d want to spread the cost of hardware over the period too?
7.
by sarjann 15d ago
What are you actually trying to get at? Had you even bothered to watch the 2nd (short) yet? My point was that he’s willing to acknowledge that he was wrong, that is what the clip shows. Also can you read minds now? ‘He's also just sayi
8.
by sarjann 15d ago
I can't find a counter to every statement they have made and not sure about your specific AGI claim but. https://www.youtube.com/shorts/QMhtTO3u61A https://www.youtube.com/watch?v=L_ueDUrkOlQ Exam
9.
by sarjann 15d ago
Not sure about being early when it comes to points about models reaching the peak of their capability. Honestly, I don't understand how he's qualified (in terms of knowledge) to make such claims.
10.
by sarjann 15d ago
Circular financing is an issue if the wealth accumulation stays in the chip -> model provider ecosystem. However it seems like the labs are making quite a lot of revenue from the chips (customers).
11.
by sarjann 15d ago
I think one important distinction is at least they have some humility to admit they were wrong. I think Ed Zitron has rarely, if ever acknowledged he was wrong.
12.
by sarjann 15d ago
I've always been suprised with the amount people take him seriously. I think he's someone who makes people who dislike AI "feel good". I don't blame someone who doesn't understand AI or have read what he's
13.
by sarjann 24d ago
I think an important part is communication, so many of these issue could be fixed by saying "Hey we're seeing our utilisation go over x% over the weekdays so we need to implement "surge usage" during this period starting
14.
by sarjann 24d ago
I guess the Riemann Hypothesis is an easy task then.
15.
by sarjann 26d ago
That's why it's a choice. If you weren't going to sign up to a site then it's just a pure degradation of experience to see links you can't use.
16.
by sarjann 26d ago
I would like to be able to whitelist some paywall sites, e.g. I have an FT subscription and would like to have that bypass the block.
17.
by sarjann 28d ago
Great news, has made low memory bandwidth model usage so much nicer.
18.
by sarjann 1mo ago
Except he tries justifying it. "Github’s servers are constantly on fire as their usage increased something like 50x due to LLM sloppers pushing large amounts of trash code" "Unless you have a emergency hotfix (you don’t), go
19.
by sarjann 1mo ago
This has to be rage bait, this is a critical piece of infrastructure for many people. Stuff going down can lead to deployments failing and as you mentioned in many cases emergency hotfix's. The idea that we should be fine with this unr
20.
by sarjann 1mo ago
You would think for a company that has close to a monopoly of developer mind share wrt git that they'd maybe put in the bair minimum to get several 9s.
21.
by sarjann 1mo ago
I think at 1000x it's probably way cheaper to do an "AWS outpost" style thing. Although not sure if it's a fundamental constraint or something that could be optimised.
22.
by sarjann 1mo ago
Introductory price seems a bit weird as it expires at the end of the year and by then it’s going to be significantly outdated.
23.
by sarjann 1mo ago
Might be easier to do cross platform.
24.
by sarjann 1mo ago
Fundamentally I always try to look at incentives. It's not that they are good or bad but that incentives favour certain behaviours. Google is incentivised to collect a bunch of data (like FB) to improve their ad serving etc. Apple does
25.
by sarjann 1mo ago
If I were to guess, they may have some "test" users (people who are involved in this) who are on the watchlist, they do their regular commutes and see if the system picks them up each time. For this system there is no requirement
26.
by sarjann 1mo ago
Google + Motorola aquisition vs Apple is a prime example.
27.
by sarjann 1mo ago
I do wonder if we need a charitable foundation that does software patents and makes them freely available just to stop cockroaches abusing patents. Another step (although governance would need to be super good for this). Would be for this c
28.
by sarjann 1mo ago
I do think some of features in their harness seem interesting (workers in separate worktrees at once), recovery from crashes seem interesting.
29.
by sarjann 1mo ago
I don't see how not? A theoretical physicist can do all the thinking they want but if they can't test an idea against nature it's not super useful.
30.
by sarjann 2mo ago
dgx spark, nvfp4 so I have spare room for KV cache (context)

More ›