cgio

Genuine question. How many big problems were solved by people who took the safe path for PhD? I would intuit there’s high correlation of unsafe PhD and big problem solvers.

Maybe they should target more ambitious results now that they can solve their other results in 15 minutes? This spirit of racing in academia is insane. Get in before the door closes.

This would that the proofs would be almost done anyway, which statistically could not be the case, or that these mathematicians were already progressing thanks to ChatGPT. Still a provenance question, but the impact of AI is unquestionable with regards to outcome.

Gödel blocked the path to axiomatic derivation of a full consistent body of mathematics as far as I understand. Mathematicians and AI are not working in these constraints but rather with these constraints.

I thought that was from the outset the intent of the Hilbert program, to automate mathematics. And mathematicians were behind it. Cannot see why they would be concerned when a different way to do the same, not subject to Gödel incompleteness, is working out. Maybe the frustration is that they were not the ones building it.

Ranking by influence. Claiming that adjacent personalities have influence that could even be compared to that of AI behemoths signing agreements with the President of the US is not going to convince anyone.

Maybe they just disappeared because they were more attracted to HS women and therefore gradually stopped reproducing in their tribes.

When two opposing forces meet, you don’t get truth from either side, you get it by critically looking at what’s left between the lines. And it doesn’t look good for EA if I go by the discussion here. Lots of pushing problematic to crazies and anecdotal altruism that is not in any way related to EA principles specifically. I don’t judge. Everyone is entitled to do what they think is best. But we can also be concerned on our side on the impact EAs actions can have on us who are not in agreement with the vagueness of it. And we can say that it is looking as much about altruism as Scientology is about science.

Any conspiracy has simple goals. Control, power, money sometimes as a means to the first two. It’s not even conspiratorial right now to assess that EA members go for these. Not all of them, but the ones that matter in this discussion.

I wouldn’t be against the co opt interpretation and I do believe people will be co opting something both with good and bad or crazy intentions. The thing is, if a fraction that co opts is so overwhelmingly outranks everyone else, then the original intent may end up the fringe. From my perspective the issue starts with the name. Why effective? The moment you introduce an optimisation problem you open the back door to biases. I would suggest blind altruism. Engineers can accept readily that random processes can optimise. Let the randomness optimise blindly and just be altruistic. You don’t have to publish an altruism balance sheet.

The smear campaign will work if that’s all that’s out there. Are they public with their beliefs/principles or does someone need to go through levels of billions valuation to know the truth.

That too… are they building a government? Because that would be a much more honest name in that case.

You may do that with honesty and just be a useful front for the heads behind the curtain. In their calculations you are still the value of 32 prawns or something.

When I talk about ineffective altruism I talk about accumulating capital and shutting everyone out of innovation in the name of regulatory protection. Nothing like a daddy to protect us sounds like altruism.

I am not familiar with their beliefs (open to learning). But what circulates is troubling. E.g. statistically optimising for future even if it means harming current can be bad dressed in an excuse. Assigning value e.g. to lives (from prawns to humans) can also be a statistical way to do harm with an excuse. Overall, if it quacks like a duck and walks like a duck…

A little bit like electricity plants. Countries will nationalise for sovereignty reasons, as China already does probably. And that’s not bad for people given current trajectory.

Can we also put a hard cap on how much ram or gpus specific entities can get?

An almost trifecta. I am almost gifted, autism and adhd ( or add, you wouldn’t call me hyperactive…). My trick was to “use” my autism to filter out rejection clues. I am still sensitive in a sense, can get obsessed over detail of something that could be perceived as a rejection, but I actively refuse parsing anything as such. People tell me I am over-confident to a fault.

Or at least an employment offer!

Then why scan them in the first place?

Even NATO calls him daddy…

The concern for them is if that small group expands. Moving back and forth limits perceived lock in. Someone who moves between oai and Anthropic will also move to e.g. an open source when it’s good enough for their purposes, or will start experimenting and combining. This is how the market will slip through the attempted control.

Is that why US failed while building CP/m, Unix a few, Dos, Windows, MacOs, OS/2,…? A successful market does not impose monopolies. The more, the merrier.

Cannabis was the same to some extent. Psychedelics too in the context of rites.

I only have Windows at work, which means less pop ups. Nevertheless, lately I get forced updates with 5 minutes “grace period” more often than I’d like. Soon will be approaching pop up frequency.

Maybe the times are changing. Went through with Adblock and vpn. No looping.

And to whom will this settlement be paid? To your random blogger or coder on GitHub?

Pretty sure that, even though the source name is changed, the article has enough detail to identify them.

I would argue using Claude in replacing a spreadsheet is potentially more of a tech debt than the spreadsheet itself. Now if you would use Claude to build a point application that performs the same standard analysis it might be better, but only marginally so.

Indeed. Some would go, some would come.