This is incorrect because LLMs don’t have the same property as that of other tools. LLMs enjoy compounding effect of intelligence which you can’t get my separating them. If what you said was right then we would see more open weight domain model. We don’t and there’s a reason why.
its correct because im using relative sizing and you are using absolute sizing, and completely ignoring the AI Bloat belief that we just need larger and larger models.
> Their models, such as Sarvam 2B and Sarvam-M, are fine-tuned for medical reasoning and symptom triage in local languages, without the need for high-end devices or constant internet. These systems can summarize patient notes, offer diagnostic guidance and even prioritize cases, functioning as low-cost, frugal AI assistants for overstretched healthcare workers.
Wow bad idea. Domain specific models simply don’t work. Ever. You should not be using some shoddy 3M model for medical purposes when you can spend just a few dollars extra and get GPT that is miles and miles better. The local language value proposition is also exaggerated.
This article keeps repeating the lie that network is hard to find in India and that local models win. This is on the face ridiculous to anyone who has been to India. Almost everyone has access to a smartphone with 4g connection. What they don’t have is the ability to afford a phone that can run a good model. Why would I as a poor farmer in India, use an extremely underpowered 3B model on my 100 dollar smartphone when I can use the free version of ChatGPT that is miles ahead in every dimension?
My 1000 dollar iPhone can barely run Gemma 4 which is hardly usable for serious questions anyway.
I do get the need for Indian ecosystem to build internal competency so that when the time comes they are prepared. But for now pursuing a distillation attack strategy like China looks better. Or have companies that specialise in integration locally - something big model companies don’t have expertise in.
Agreed, and also now any actions people don’t personally agree with that have a negative effect on someone (which is everything) can be labeled “structural violence”, so they can claim to have a moral justification for physical violence. It’s disgusting and breathtakingly stupid.
Disregarding the fact that Bryan operates oxide a company that has multiple investors and customers (id say this proves valuable knowledge) the crazier fact is that people think html is useless knowledge.
React USES html. Understanding html is core to understanding react. React does not in anyway devalue html in the same way that driving automatic devalues driving manual
Go to Facebook.com and right click view source and tell me html is not being devalued. No person who wants to write aesthetic html would write that stuff.
Tbh this is like looking at any binary's assembly code and complaining it's "unasthetic".
Whatever blob of HTML the browser is offered is just the end result of multiple "compilation" steps. Nobody has spent a single iota of time thinking about whether it's pretty.
Just like opening any binary produces inelegant assembler.
When it matters it matters. Even in facebooks case they made react fit for their use case. You think the react devs didn’t understand html? Do you think quality frontends can be written without any understanding of html?
Like the article says we’ve moved an abstraction up. That does not make the html knowledge useless
The content of his talks is very good, but his habit of SHOUTING all the time just grates my brain the wrong way. I can literally see the veins in his neck bulging as he screams out his presentation.
I would seriously consider if you've developed an imaginary caricature in your mind that you apply to people you don't know. Further, I would consider if any living person actually lives up to it.
On the one hand, I admire (at some level) you sticking to your guns here, willing to take on all comers. On the other, though, I don't entirely understand the inference that you're drawing from the piece; what, exactly, is getting commoditized?
What he prides himself in (in this context) is craft, which LLM use probably can enable, but definitely isn't commoditized by the kind of vibe coding that Garry Tan is doing.
I feel like Europe needs to remove barriers and let people do things freely rather than stuff like "Empower AI students". Ambitious people will naturally find a way to get stuff done and you just have to allow it to happen and not get in the way. At least "EU AI talent visa" sounds like it can work by removing barriers to relocate.
I don't see the AI talent visa as a serious suggestion. There's already the Blue Card, which is pretty easy for software engineers to obtain; for an AI talent visa to be anything special, it would have to loosen the already fairly lax requirements. To manage to persuade the whole EU to create a special visa for an emerging field that doesn't even have commonly-agreed-upon qualifications would be a unrealistic feat when many EU member states are looking to close down immigration opportunities, not create them.
META used to hire in Netherlands until it stopped and left. I wondered why and a few times I heard that it was because it was hard to be dynamic in the country with the stubborn labour laws. The anecdote confirms my own bias but there seems to be not much mention on encouraging risk taking allowing dynamic entrepreneurship in this playbook, which leads me to believe this is a non issue?
> it was hard to be dynamic in the country with the stubborn labour laws
It's hard to read this as anything other than "It's hard to treat your employees like shit when countries have strong labor laws and that just won't do for american megacorps".
While I don't think Mythos is so powerful that it justifies containment permanently - I wonder how it might work when there is such an AI that can justify containment.
What if this new model can start proving Millenial problems and provide insights in other fields that was not possible before?
My intuition says that a model that is as good will also be equally well aligned -- but it is still highly risky to give it to the general public because all you need is one jailbreak from bad actors.
At that point I think society would change so dramatically that "access to general public" would be a non issue. Rather, time would be spent on making abundance happen - you might think of the political struggles, economics and new ventures.
Its a bit sad that democratised access is not provided because of negative sum possibilities like cybersecurity.
reply