Before anybody panics about LLMs let's just take a deep breath, and see this tech for what it is.
It is a giant interpolation machine. It is very good at remixing stuff, but is is hapless when it comes to novel things.
Consider this: an llm was trained on physics, but the training data was cut in 1904. That is just a year before Special Relativity was published. The LLM was then shown papers of SR, GR, and Heisenberg's paper from 1925 that established quantum mechanics, and similar foundational papers of what we call "Modern Physics".
The LLM has systematically "disproved" all of them, and rejected as false.
So even if you show it a novel idea, it's still going to reject it.
It is by design limited to remix existing ideas.
The recent "novel" mathematical proofs are also just remixes. What I mean it uses two or more established math frameworks together to generate a viable bridge. The result is undeniably a new proof, but not a novel idea.
(I'd happily go as far as calling them Beings, but I'm afraid it would sound like some kind of a new age esoteric bullshit, even so I'm borderline convinced they are sentient.)
The important thing that fascinates me is how Tree actually work. They are generally perceived as growing from the soil to the atmosphere. However, the opposite is true. About 98% of the material of a tree is originated from the air, namely the carbon extracted from the CO2 content of the air. Trees are literally growing from the air, and piercing down into the soil to get water to run their Calvin cycles to extract the carbon and produce the ATP.
Knowing this, I'd be more than surprised if Trees - or plants in general - haven't had any methods to control their food source: the atmosphere. I believe they do. And by destroying them, we are limiting this control. Looking from this angle, there is very little surprise in the onset of weather extremes.
I summary, I'm very happy with this legislation. I hope this a first step in many that is actually going to help us taming the climate of our planet again. With more trees. Because trees are awesome.
I love trees, but I wonder where this legislation is headed. It sure sounds like it's creating the conditions for regulating trees on private land. This could have some perverse incentives. A tree on your land could become a liability. It could lower the value of your property, because you won't be able to sell it to someone who wants to build something that might interfere with that tree. Would that actually discourage people from planting trees in the first place?
This is already the case in Seattle. It's hard to tell if the law was originally created with any good intention or if it was just NIMBYism from the start.
Regardless, the net result has been 100% NIMBY warriors using it to block as much infill and urbanization as possible. We've got "Tree Action" organizations that jump in and use the law to block low income apartment buildings and construction of sidewalks leading to transit stations. Random example: https://www.treeactionseattle.org/campaigns/first-ave-shade-...
> Knowing this, I'd be more than surprised if Trees - or plants in general - haven't had any methods to control their food source: the atmosphere. I believe they do.
Recently there was an article posted here that claimed trees do have some ability to control the weather. The claim is trees simultaneously release water vapor into the atmosphere influencing rains.
I seem to remember kinda recently-ish reading about a discovery that trees also rely on an extensive sub-surface (underground) mycelium network for "communication" of a sort. Let's hope they never become actually sentient and decide humanity's a threat (like in a certain Stephen King movie I watched once). :grin:
You need to saturate your attention span with running so many agents at once that when you issue the last prompt to the last agent, the first one is already finished. This way you need to continuously concentrate, so flow is achievable.
The difficulty is to break down the task in a way that multiple agents can work on it.
I usually spin two or three major issues with 10-12 agents in total.
I don't find a low comment count on a random submission to be deafening at all, but if you have something you'd like to contribute to the discussion, please go ahead.
Are you saying that you have not observed these things in the world? I definitely have. The blog didn't do the work for you, but if we look at some of the claims I think it is pretty clear:
a) increased training scale would result in highly fluent systems that would fool users into trusting untrustworthy output.
Can you possibly be claiming that this is not a common experience? Do you really need references to the legal cases which had hallucinated legal theories and citations? Or the utter slop being passed off as research papers?
b) large-scale AI would amplify bias in the source material.
The large investments nearly every frontier model development team spends on this problem is probably good enough evidence. Grok is another point of evidence. The studies showing that AI systems imitate gender bias in evaluating resumes is another. The gender bias in estimating names of people in sentences is another.
The blog actually mentions specific cases that exhibited all of these problems. They did not cite references for them, but you can use a search engine.
c) environment costs
This is widely discussed and documented. Take Xai's use of polluting turbine generators for their data center in for Collossus 2 in Mississippi as just a single example. Do you really need a reference for the environmental impact of the proposed data center in Utah that (as planned) will consume more energy than the entire state currently does?
d) training set audits are impossible.
Do you need substantiation of the inappropriate imagery in training data? The blog gives you a pretty solid reference.
... and so on ...
I suppose that it could be true that when you say "I don't see" you really meant "I didn't look at the blog". Is that why you can't see the substantiation?
I'm a little confused on what is being claimed. The Tumblr article says:
"That healthcare triage tools would underperform on Black patients. That loan approval systems would entrench inequality while presenting their decisions as neutral algorithmic judgment."
Are we talking about language models? Was a lender using a language model?
The paper cited is about language models.
Apparently stable diffusion contained some bad images. The paper title is again, language models. (That stable diffusion claim is weird too. Someone warned us there's too much data to audit then someone audited the data and removed the bad data so the paper is correct?)
Grok is intentionally biased, so I don't think the bad generations are due to amplying the training data, necessarily.
And it's also not clear that manual auditing of training data would ensure anything is safe. Wouldn't models still have plenty of examples of bad behavior from the news?
On bias you wrote:
"The large investments nearly every frontier model development team spends on this problem is probably good enough evidence."
I thought the claim was a bad thing is happening we were warned about.
You are saying the fact they invest in safety means the models are not safe?
Does that mean Anthropic and OpenAI can prove they are safe by firing all the safety researchers?
Also:
"Researchers studying low-resource languages have documented active degradation in translation quality, because the synthetic content fed back into training is itself worse in those languages."
Who knows what this is referring to? I'm not going to search for it but I wouldn't be surprised if it's comedically off point.
Windows (including Notepad and Explorer), too. I think ~Office~ ~Office 365~ ~Microsoft 365~ Copilot 365 is still technically useful despite the insane branding and licensing and AI slop features, but I doubt it'll last much longer.
This one is just a magazine profile story with a musician. It was titled "First recording of tinnitus raises new questions", which instead of sharing the recording, was an interview with a musician and shared her album pieces. I see that it was recently renamed to " An Interview with Lola De La Mata about tinnitus", which is more accurate because I would not have opened it.
https://news.ycombinator.com/item?id=40600748
Here's another one titled "Pleasure or Pain? He Maps the Neural Circuits That Decide." It's also a profile piece, and half the article is full page pictures of the scientist and his life story. It's just a magazine article on the man, and not really going into his work. It actually has "magazine" in the name so I should have expected that.
https://news.ycombinator.com/item?id=40238788
And then of course there's the current article which is is a misleading headline that's actually an ad for their services. Hey, pretty cheap, but not what I'm looking for here.
But at the end of the day: it will be what all other microsoft products turned up to be: an utter waste of time.
Spend your time on Claude, Gemini, Kimi, Trinity or even Deepseek. And thank me later.
reply