What I like is it's still hard, but you can do things like "prove this using game theory" or "find the optimal value for this and a proof" without having to know game theory or deep math really
There used to be a lot of research into using deep NNs to train decision trees, which are themselves much less of a black box and can actually be reasoned about. I wonder where that all went?
Very ironic for the billionaire to be openly replacing himself with AI, I suppose he believes his job is easy enough that an LLM can do it, so we definitely don't need him
It's all just really genius marketing. In 6 months Mythos will be nothing special, but right now everyone is being manipulated into fearing its release, as a marketing ploy.
This is the same reason AI founders perennially worry in public that they have created AGI...
Yeah perfect example, the main thing I _would_ use multiple agents on is optimizing/benchmarking code, but for that you specifically can't use worktree, you need one agent per machine or they'll taint each other's benchmarks
It's inherently free data. You have to be able to trade against it, and to trade against it you need to see it _before_ you give them money, so they can basically never charge for it. When they sell it to people, it's just laziness, they could scrape it themselves.
edit: apparently they do, that's even more wild lol
> It's not some kind of magical money multiplication machine
Right, it's not even consistent, the manipulation can go either way, at random, but what is sure is the higher the liquidity, the higher the motivation becomes to tamper
Especially when the "prediction" part isn't even part of the product, it is an accidental byproduct. No one pays for access to the predictions.
And they aren't predictions, they are more like "outcome-shaping" markets, since the more liquidity that gets dumped on a particular outcome, the more motivation there is to tamper with the real-world outcome, and at a certain point it will just always happen if it is billions of dollars.
The higher the liquidity involved, the less likely the real world outcome ends up being the same as it would have been if the prediction market had never existed. Very messy.
Polymarket being an incompetent cabal that doesn't deserve to have a license is so passe, so ingrained in the standard recurring narrative that comes up whenever prediction markets are mentioned, and so universally agreed upon at this point that it frankly isn't that interesting to a lot of people. We all agree with it already, there is no one left to convince, certainly not on this site, better to talk about more interesting things that are nevertheless related. It's, well, boring, and this is exactly why a comment like this wasn't the top comment. I'm not saying a comment like this wouldn't be true, it would, it just isn't going to be a top comment, and that fact is unsurprising. I do share your sentiment though, that it would have been nice if human attention worked slightly differently here, but it does not.
> Hand over the names of the large bettors on that side of the market to the Israeli police
Woah there, didn't see this coming to be honest. But, I mean we already trust Israel to spend our tax dollars on genocide, so heck, why not, they'll do a great job!
> If Polymarket (and its competitors) cannot fix this, then it's time to get tough. Maybe it's time to actually crack down on Americans using VPNs
First of all, why is it time to get tough on limiting human rights, why is your freedom of speech and freedom of privacy the first thing on the chopping block, rather than Polymarket's license to operate, which is a much more obvious target for all of our disdain.
And also, it's physically impossible to crack down on VPNs 100%, especially within the constitutional framework we have in the US, and the technical reality of the modern internet. Keep in mind the US Government itself invented TOR to help destabilize and leak information out of autocratic regimes that have implemented sweeping censorship and blocking of VPNs (you know, the sort of autocratic regime you apparently want to start here in the US given your comments on the matter).
From a technical perspective, with SSL being standard for most web traffic it is utterly trivial to set up VPNs that hide in all kinds of ways that look completely innocuous and are unblockable. You can even hide exfiltration traffic in DNS queries, and I even got to see this a few times in practice when I still worked for the DoD back in the day. People will go to all kinds of lengths to get information in and out of a closed system and stopping that is like trying to catch air with your hands, you will always just end up hurting/restricting the average citizen and the bad actor who wants to slip through will still slip through.
Such a ban or KYC program is also fundamentally against every notion of privacy and freedom enshrined by the constitution and even more so by the founding spirit and values of the internet and reputable tech circles that care about privacy and freedom. You sound just as bad as Zuckerberg and Palantir trying to push KYC on Linux right now to help build a better mass-surveillance state for their fascist overlords. You seem to be the sort who would be extremely supportive of that project, though I sincerely hope that is not the case.
> we have extradition with Israel
For now. Once the Boomers finish their reverse mortgages and are finally out of the equation we'll hopefully cut all funding, sanction Israel, and label them a terroristic genocidal state, which they are. As usual, moral progress is always delayed by rich old men refusing to let go. Ethnostates are also extremely problematic in their own right, never mind genocidal ones. Certainly not morally praiseworthy, certainly not good "ally" material. Should be sanctioned in the very very least, definitely not subsidized.
And the funniest part is I agree with you that Polymarket is bad, which is your underlying point that you care so strongly about throughout this thread, but your arrogance and sudden undeserved spitefulness, in response to completely innocuous, imaginative, on-topic comments, just comes off as really bitter and makes you hard to talk to and even harder to agree with. Like you really have an axe to grind and I think it is rooted in some sort of deep-seated fear that everyone around you is as evil and misanthropic as you claim to be through your espoused values.
> because it was the only way I could eke out something remotely relevant to the article from your original comment.
Remotely relevant? It's not like I'm talking about cats, I'm talking about prediction markets, which is the topic of the article. You would have to be extremely uncharitable to not see a holistic discussion about other problems prediction markets face and possible solutions to those problems as relevant. These topics very seldom come up on HN so a mere mention of prediction markets can and should be a very good excuse to have many side threads on the topic that have been waiting to happen.
It's also fundamentally something you can't really do properly in crypto. You need a central authority to decide if "thing has happened". There's no way to properly incentivize accuracy in cases where the majority of stake benefits from an inaccuracy.