It's not just America. The main secret is out of the bag. If it wasn't Anthropic, it would be another company/nation state. Sure they could obtain, and with not.money or leverage, complain about data centers at local rallies, or they can be in the game, and hopefully steer it. It's going to happen with or without any one company or country. The secret it out, and it's unstoppable without complete societal breakdown..So either you advocate for the end of civilization, or you hope that you can help steer the emergence of super intelligence into something not wholly terrible.
Personally, I don't see much hope, even if there wasn't such a thing as AI. The power to destroy is always easier than the power to create, and as our power grows, the differential grows, until at some point, containment is no longer possible
I'll mention again the nuclear analogy. It is, believe it or not, possible for great powers, and even adversary great powers, to agree to limit the development and proliferation of dangerous technologies.
> The main secret is out of the bag.
This is not something you can do in a shed with a handful of GPUs just because you know "the main secret". To build something like Mythos you need tens of billions of dollars, massive amounts of power, enormous buildings filled with specialized bleeding edge computer chips that are made by (optimistically) a handful of companies with deep government ties. You need free access to all the intellectual property that humans have created and posted openly on the internet. You need all of this at each step, and to take each next step you (or somebody) needs to have taken the previous one.
For now, there are a million ways for a government to pump the brakes on this cycle.
Yeah, I think the nuclear analogy fails, honestly. The bomb does one thing: Destroy. AI can build, selectively infect, selectively manipulate. It is a vastly useful tool.
What is the difference? It's easier to make money with the AI you get at each incremental step toward potentially destroying human civilization (though, of course, it's debatable whether these companies really are making money as such).
So what? You are implicitly arguing that human civilization will be unable to resist engaging in a large-scale, coordinated effort to destroy itself, just to make a few bucks along the way. Is this true? I don't know. The point of the nuclear analogy is that we have previously shown that we can, under certain conditions, put the eschaton back on the shelf for some period of time, despite very real pressure to take more incremental steps toward doom. "But AI can write code" is not a refutation of the possibility that we could take a more measured approach to AI development.
I never claimed to be precise, ha ha. But let's not lose sight that each of the great powers agreed to limits on the number of nukes only after making enough to collectively destroy each other several times over. They stopped exactly when building more did not gain them any more security or advantage over their adversaries. Not a moment before that.
There may or may not be such a point with AI: A point at which ever smarter machines provides no marginal benefit to security. If that happens, I do expect agreements, should any one of us still exist.
> It is, believe it or not, possible for great powers, and even adversary great powers, to agree to limit the development and proliferation of dangerous technologies.
You were literally just criticizing Anthropic as disingenuous for begging for this. Or is your position that people other than those near the front of the race can agree to limit development? And if so: provide evidence.
Note also a key ingredient that makes nuclear non-proliferation possible is that they're pretty much useless weapons. There is no smaller order nuke that's dramatically more useful than a large conventional weapon. That's not true of AI models, which appear to be monotonically useful as they become more powerful.
Fishy to me: They report 0 refusals on security tasks, yet I can't even get it to code a task involving choosing the best mixed model, extracting BLUPs and propagating uncertainties.
California has 4 seasons of growing weather, good soil, and conditions that help against rot. The lack of water is exactly why it's good for growing stuff, if you can get the water. Excess water ruins crops all the time.
I was a bit disappointed that it refused to use Fable to help check whether I was propagating uncertainty from BLUPs in my random effects model up to the subsequent group level analysis in a maturational coupling analysis of brain data.
I guess brains and random effects blew its lid.
I wish that the tutorial went just one more step. It presents a one dimensional perceptron. But most perceptrons are multi-input.
Adapting the article's 1D perceptron to three-input, for example:
import random
learning_rate = 0.1
EPOCHS = 50
NUM_INPUTS = 3
weights = [random.uniform(-1, 1) for _ in range(NUM_INPUTS)]
bias = random.uniform(-1, 1)
data = []
for _ in range(100):
inputs = [random.uniform(-1, 1) for _ in range(NUM_INPUTS)]
result = sum(inputs) > 0
data.append((inputs, result))
for epoch in range(EPOCHS):
for inputs, result in data:
weighted_sum = bias
for i in range(NUM_INPUTS):
weighted_sum += inputs[i] * weights[i]
prediction = weighted_sum > 0
if prediction != result:
error = int(result) - int(prediction)
for i in range(NUM_INPUTS):
weights[i] += learning_rate * error * inputs[i]
bias += learning_rate * error
print(f"Final weights: {[round(w, 3) for w in weights]}")
print(f"Final bias: {round(bias, 3)}")
First, I would not say "debunked" is the most appropriate term. The studies were not shown to be false, but the article highlighted that come doubts have been cast due to potential contamination confounds. The letter (Challenges in studying microplastics in human brain https://www.nature.com/articles/s41591-025-04045-3) will get integrated into methods research. See here for current citations: (https://scholar.google.com/scholar?start=0&hl=en&as_sdt=2005...). Sciencing is actually harder than it looks from the outside, and therefore research fields goes through iterations of refinement. As the article you linked to included this quote: (Prof Lamoree “It’s still a super-immature field and there’s not many labs that can do [these analyses well]. When it comes to solid tissue samples tissues, then the difficulty is they are usually taken in an operating theatre that’s full of plastic.”).
Second, I do not think anyone is claiming that microplastic pollution is not ubiquitous, because that is obvious. That microplastics get consumed is also probably not that controversial. The extent to which microplastics get consumed but do not exit the other end of the pipe is an empirical question that has methodological challenges.
Third: I think there are some subtleties here involving the size of plastic particles. Microplastics is a catch-all term these days, but a more formal definition puts microplastics at plastic particles that range from 5 mm to 1 μm (micrometer), while nanoplasticsare 1 μm down to 1 nm (nanometer). micro- and nano- plastics can require different techniques to detect.
reply