Upvoted: I was going to write this exact reply. Everyone is a bit tired of hearing about AI, but it's advancing fast and it's the most important thing right now. As Demis says, if we get to ASI, we can solve all other issues with it.
Actually, it (LLM models) has been advancing slower for a while. This is pretty much in the mainstream. And the current approach is facing all sorts of obstacles that are not exclusively technological. The compute needs are no longer sustainable. Financially, big labs are drowning in debt and backlash against data centres has grown so much that it has become an electoral topic in at least one major country.
> if we get to ASI, we can solve all other issues with it
Jesus Christ this has to be bait. Added to the favourites to laugh about it when the lunacy and hype has died down in 5 years and the trough of disillusionment has truly settled in.
Or to laugh about it when ASI actually has arrived and we're waiting in line to be turned into biofuel for the Greater Good.
Everybody is literally pinning their hopes that God itself comes out of machine to save them from the exponentially accelerating pile of tech debt, and they're not even half joking about it. This is not engineering, this is utter madness.
I'd say it is engineering. There's a Moore's law like steady progress where the effective IQ goes up each year which will result in neither troughs nor biofuel but just smart AI.
Given human nature I also think it'll be kind of undramatic. Like when jet aviation transformed things people moan about check-in queues and maybe with ASI people will probably moan their robot butler is only 200 IQ when their neighbours is 205.
Shit we’re onto thinking ASI is happening soon? I guess people hyped up AGI so much that it’s no longer thrilling to talk about despite not getting to that point either.
why exactly do we want ASI, and further, if we DID, why on Earth would we want it controlled by the likes of Sam Altman, Peter Thiel, Elon Musk, Marc Andreessen, and Larry Ellison?
GLM-5.3 is further proof that all >1T models are currently undertrained. I was looking at inteligence density ( https://www.pasteboard.co/6q2-5f92mtj9.png ) from recent open models (where parameters sizes are known) and taking DS-v4-flash as upper limit GLM-5.x can 3x its performance.
How does predicting a typhoon prevent billions in damage?
It could save thousands of lives because people can be evacuated if you can predict a few hours or a day further ahead, or the path more accurately. You can save some damage by moving ships and vehicles.
But you can't evacuate buildings or infrastructure.
You can board up a helluva lot more stuff in a week than in two days. Crucially, you can move more of the most expensive stuff out of the storm surge zone, which is where the biggest damage happens and try to flood proof more of the things which can’t be moved.
> You can board up a helluva lot more stuff in a week than in two days
TFA says they might be able to predict one extra day ahead (three days instead of two). No prediction system will ever give you a week's notice on a typhoon.
Actually it doesn't even say that - they claim to have the same "accuracy" at 3 days that older methods have at 2 days. What does that actually get you? Were the older models so much less accurate at 3 days (compared to 2) that it prevented evacuation of key areas? Looking at the paper, it doesn't really seem like this can be answered yet because there's not enough data over a long enough time.
Keep in mind, DeepMind has a very well documented history of releasing enormously hyped up PR pieces with grandiose claims that are never backed up in real world usage, or are simply lies.
I wish there was a way that your comment could be pinned.
The context is so important here and radically re-frames the impact of GDM's results. Folks need to understand that with modern forecasting tools, we anticipate tropical cyclones to develop 5-10 days before they ever threaten landfall. The "2-day" vs "3-day" improvement in forecast skill is better interpreted as a modest reduction in forecast uncertainty - the "cone" on the hurricane track map gets a little narrower.
It's not like there's a "literal extra day" of preparation time for folks who may be impacted by the storm. They get the same amount of time they always have. Nothing actually changes on-the-ground for really any consumer of hurricane forecast data anywhere in the world.
And that's not a sleight against GDM. It's just a simple statement of how good contemporary weather forecasting is, and how good it was before AI forecast models came onto the scene some 5 years ago.
I know this is uncharitable and I am wrong but I am having trouble coming up with concrete scenarios where you die with 2 days notice but survive with 3. I am nonethless a believer that more accurate forecasting has value.
You can walk 50km in those extra 24 hours which will save your life if you’re in a storm surge area without other means of transportation, which is easily the case when everyone else is evacuating alongside you.
Could you imagine a scenario where from warning to complete evacuation takes more than two days? Evacuating a whole area is a hard task, particularly once you start looking at more complex problems (elderly, prisons, hospitals).
I feel like the details of this are highly dependent on the confidence of the warning; moving large numbers of people (particularly elderly) will result in some deaths regardless. I guess more time to do it should help though.
It's not uncharitable, because the system doesn't actually claim to give you an extra day of notice. It would be more accurate to say: "the model can reach a given level of forecast accuracy roughly a day farther in advance". So your question becomes: are there scenarios where I am 80% sure this is a Cat 5 hurricane 3 days before, where I would die if I was only 65% sure it was Cat 5 on that same day? The answer is - probably not, because even in the example used in the PR article, the NHC was already issuing strong early guidance 5 days before landfall.
I don't think the amount of notice is as important as "this totally is for sure going to wreck you" is. We generally know something is going to hit somewhere at sometime. Which isn't specific enough for everyone to act on
I've always assumed that insurance and/or government departments that would spend money due to storms would be the ones funneling money to these sorts of efforts. It's not exactly something you can easily sell directly to individuals who would benefit. It would be pretty dystopian for them to sell subscriptions for an extra 24 hrs notice on the next typhoon :P
It is not a risk is a fact - people decompiling Claude Code have found many times that it has code branchs to detect it is being used in Chinese timezone and locale.
Zhipu AI is founded by a superstar Tsinghua professor, did an IPO in January (Hong Kong stock exchange) hired half it's past research lab and it's stock is >10x since. This is not a "just distill Claude" thing.
