Nvidia Is Simulating a Copy of the Earth
futurism.com
futurism.com
Because we do have weather balloons and weather satellites and weather stations in variant heights and situations.
Countries the size of Poland go without for at least a decent chunk of the 8760 hours in a year.
Weather balloons are extremely sparce.
> The Aeolus mission objectives were to provide accurate global measurements of winds from the surface up to 30 km
It's been proven for decades, the big problem with weather simulation is that the dynamics are chaotic. They continue to be, even as we collect more data from the surface and above. Our forecasts improve, slowly (by a day per decade or so, historically), but simulation cannot be accurate and precise in the same sense.
A good introduction is "Weather Prediction: What everyone needs to know"
In that show, policy makers use a computer simulation of the world to quantify the limits that they set on emissions, and all of the nations in the world come together to make it happen and look back on this age as an important lesson.
I believe there's a low-res VHS rip of it available somewhere if you search, which gives the whole thing an eerie feeling of discovering an alternate-reality documentary.
Whereas if they said "Nvidia is creating a highly detailed climate simulation", very few people would read it.
For the latter, emphasis on the model as a platform vs just a simulation, makes sense. I.e. a “digital twin’ meaning a rich flexible model capable of being the basis of many different simulation scenarios, not limited to a designed in class of simulations.
> coming soon
> coming soon
Answered my questions, thanks
Most prominently reminds me of the EU’s Destination Earth
https://www.ecmwf.int/en/about/what-we-do/environmental-serv... - another digital twin of Earth climate
NASA has another project https://esto.nasa.gov/earth-system-digital-twin/
If it's not perfect in every way, it's bad.
Let me guess, it's not green enough for you?
Imagine if climate change advocates understood the fate of the Earth's climate is determined overwhelmingly by poor people in countries that don't care to be lectured by wealthy Western elites because they're poor. This hardware has a real possibility of enabling breakthroughs across all sectors of the world economy, and those poor people benefit the most from those advances. And since the activists refuse to grapple with this problem, I'm starting to be convinced they're just sociopathic maniacs or deeply racist lunatics. The thought of seeing anyone they deem below them rising to their status makes them seethe.
The argument is based on evidence and reason, not emotions.
> then it’s predicated on ...
This isn't a hypothetical scenario -- these people overwhelmingly use fossil fuels rather than green energy.
> It also ignores that the wealthy have the resources ...
Nobody knows whether an astroid, nuclear war, or a super volcano eruption poses as much, if not more of an existential risk than climate change. And given multiple existential risks to the Earth with finite resources, we cannot fall to silly appeals to emotion when allocating these resources. If climate change activists were honest, they would be arguing for why climate change is more of an existential risk than the other existential risks to the planet. Instead, they prognosticate the end of the world rely entirely on sophistry to get their message across.
We can 100% ensure a climate disaster, or nearly 100% mitigate the worst case.
The cost benefit analysis is clear. It costs less to mitigate climate change, and there are huge political and economic gains from removing our dependence on oil, rather than suffer unmitigated climate change.
The hard part is coordination. We have the resources and economic incentives as a whole to take the best path. But our political and economic systems are not as sophisticated, in terms of translating the benefits of change to individual actors, as the problem requires.
Externalized damage is profitable to powerful corporations, and even customers, in the short run. But we need political systems to more reliably set limits, AND provide the incentives, that would result in CFO’s at the most relevant corporations welcoming the change too.
To a first approximation, the transition from fossil fuels by fossil fuel companies will be set by those companies’ CFOs. Make it less profitable to stay with fossil fuels, and make it so profitable for energy companies to supply other forms of energy, that those CFO’s get on board and things will change very fast.
Share the economic value of a change with the most important players, as apposed to fighting them, and it will happen. Pragmatism wins.
Sounds worse than sociopathy.
High minded platitudes not physical evidence is the preference?
May as well go back to religion.
Will go with my current living experience, steeped in modern physics and math, over a long dead philosopher who had never encountered a light bulb
The past was wrong as much as right, and generally less civil than now. I don’t really see the necessity in relying on it.
Nobody knows whether an astroid, nuclear war, or a super volcano eruption poses as much, if not more of an existential risk than climate change.
It's probably going to come down to:
1. How accurate is it?
and:
2: Are the people using it for things that it suits, vs things that it doesn't?
For example, if it turns out to be decently accurate at predicting weather (say) "3 days from now" while being seriously wrong at timescales of 6 months, then only using for weather forecasts for the next few days seems good.
But some people tend to believe anything an "AI" computer says regardless, so we'd need to make sure those kinds of people aren't getting it adopted where its harmful.
Simply waving the upsides around and shaming anyone who questions the cost is an age-old posture that has helped bring our society to the brink of disaster.
- We don't have a 100% renewable energy regime (and we are very far from it)
- Even if we did it would be possible (and likely) for bad actors to arbitrage cheap energy for yet-more-harmful use cases
- As we grow the portion of grid demand that is served by green, renewable energy, the price of coal and natural gas is being driven down
- Until we have 100% deployment, one company's "green" energy consumption will simply drive the less-well-capitalized to cheaper, not-green energy sources
Cheaper, more pervasive energy will be used efficiently in the ways that are enabled by our legal framework. If we aren't critical of those usages, there is nothing to say that abuses won't become exponentially more conspicuous even as we adopt more and more "clean" energy. And in fact, that's what we see happening in the world today. Look no further than:
- Bitcoin mines exploiting cheap spot power and demand response policies, and burning natural gas directly from the well because the speculative price of crypto is more appealing than selling the same gas on the market
- Generative AI in general, but especially when you see hundreds of thousands of GPUs deployed without so much as a business case (Stability AI, Inflection, probably many others surfacing in the weeks to come)
Ah ok
I'm assuming this model is in the same vein as ECMWF, GFS, ICON, CMC, and NAVGEM.
Imagine if the NOAA or an academic lab released such a model called it "Earth 2.0" or "Earth's digital twin". It would just feel... like hubris? Not sure why (or how) Nvidia gets away with such messaging.
In a parallel timeline the smartest minds in the world are fixing climate and medical problems, further the limits of human knowledge and exploration, or even just making fun things to play with.
In this world they are trying to outdo each other in brainwashing people to improve the profits of companies larger than most countries.