Haha that's not what the post (or the post it links to) says. Every CS student should know there's no free lunch in search and optimization. There's tradeoffs between random search, evolutionary algorithms, and convex optimization. https://en.wikipedia.org/wiki/No_free_lunch_in_search_and_op...
There's an AI "smell" to things that are generated. Why is that? Mode collapse is impossible to see from a small number of samples. Are we mode collapsing society? How would we know if we were?
Also, will computers surpass humans has such an implicit bias in it. Have humans surpassed ants? Have ants surpassed rocks? Have jet planes surpassed teletubbies?
> Every CS student should know there's no free lunch in search and optimization
The no free lunch theorem is so absurdly limited because of the constraints that it's IMO a tautology and fundamentally irrelevant outside of exceptionally tiny areas. You can't have one search algorithm that's better than others on average when searching entirely random things with no structure? 1. Yes, obviously. Nice to have a formulation but it's not exactly a surprise and 2. That's not what we deal with in the real world.
The bar isn't massive benefits for the world, the MacBook Neo is great! If there was a new company that builds MacBook Neos, that's a great company. They build something real and sell it for more than it costs to make, no strings attached.
The problem with Apple comes down to the App Store, the forced 30%, and all the apps that just don't get built cause of Apple. This is rent seeking, and this is evil.
If you don't want a MacBook Neo, don't buy one and it doesn't affect your life. But the App Store affects your life whether you own an iPhone or not. It affects the direction of the world. And that's where the rent seeking problem is.
Not at all. OpenAI / Anthropic are producing tons of surplus value right now! Not to mention how great the Chinese open source LLMs are. And Apple's hardware division has always been fine.
Apple's 30% tax for payments in apps is the ultimate rent seeking example though. Want to install your own apps, lol you can't. And if big AI companies follow in the steps of Google/Facebook it's bad for everyone. Let's recognize it and prevent it from happening this time.
Building tools and services to reduce hassle and friction for others is great. However, what often happens is that you end up creating and building a moat around that hassle. Think about how companies like TurboTax lobby the government to not build electronic tax filing stuff.
Cory Doctorow explains the dynamics well in Enshittification. First they turn against their users, then their business partners, then their employees. The layoffs you are seeing are just stage 3 enshittification. If you work at a company like this, my advice is to quit ASAP. At least then you leave on your own terms.
Indeed! He specifically cites Yanis Varoufakiss, dubbing this technofeudalism. If I may quote Doctorow a bit:
>Varoufakis defines capitalism as a system designed to preference profit over rent.
[snip]
>The feudal era wasn't defined by the absence of profits--rather, what made feudalism "feudal" was the triumph of rent over profits. When the interest of rentiers conflicted with the aspirations of capitalists, the rentiers won. Likewise, the defining characteristic of the capitalist era was not the abolition of rents, but rather the triumph of profits. When capitalism's philosopher-theorists lionized "free markets," they didn't mean "markets that were free from regulation," they meant "markets that were free from rents."
[snip]
>This is Varoufakiss technofeudalism exemplified. It's an economic system in which the majority of value is being captured by people who own stuff, at the expense of people who do stuff.... The fight between technofeudalism and technocapitalism is a fight over whether the landlord or the café owner takes the value that's created by the barista.
i'm fairly certain Cory Doctorow does not understand the economics of Enshittification.
companies subsidise their products so that exploration of these products is more feasible due to lower initial costs for the end consumers. the initial consumers don't pay the full price but they are borne by the later consumers once the exploration is done and they have knowledge about that market and business.
Cory Doctorow also probably confuses democratisation and enshittifaction - its usually the case that products get cheaper by also marginally reducing the quality. we get cheap goods from China but that's not enshittification - that's just efficiency. as a consumer I'm happy I have the option of paying low prices for products.
i wouldn't take this person too seriously because it looks like they don't understand the larger picture
What are you talking about. Cory literally coined the term to describe this phenomena. He is not confused by the idea of cheaper products with wider appeal. He takes issue with vendor lock-in that is weaponized first against the end-user, then against paying customers, and finally against investors themselves. This is first and foremost a criticism of online products and platforms, not mass-produced gadgets from China.
There's no such thing as "insider trading" on prediction markets. Are you telling me people with secret knowledge profited by making correct predictions? That's the whole point of prediction markets! Provide accurate information = get paid. It doesn't matter where the information came from, as long as it's correct.
If you hold confidential information, using it to bet on prediction markets could arguably violate secrecy rules since the market effectively receives the information. If you are not bound by confidentiality, participating should be fine.
Note that this isn't about making the market "fair" for other participants but about ensuring confidential information remains confidential (especially relevant in the context of national defense).
If they provided the information, "Hey I am bob in the pentagon and we are doing war soon", it wouldnt be insider trading.
But all they are doing is updating a financial instrument that suggests an increased likelihood, and getting massive bank for not providing the information that would make everyone else bet the same way.
>But all they are doing is updating a financial instrument that suggests an increased likelihood, and getting massive bank for not providing the information that would make everyone else bet the same way.
That's the incorrect way of thinking about this, at least according to how US insider trading laws work. If a hedge fund has reason to believe oil prices will spike due to some secret info (eg. they paid some intern to camp out at US airbases and spot outgoing flights), and then they made massive bank on that trade, that's not insider trading. It's not a crime to hoard material nonpublic information and trade on it (ie. "updating a financial instrument ... and getting massive bank for not providing the information that would make everyone else bet the same way"). Now, if they paid off some guy inside the base, that might be breaking a bunch of laws around national security, but still not insider trading.
That’s 100% insider trading. If you use material non-public (including confidential) information to perform the trade it’s illegal. Paying someone to provide confidential info is still insider trading. Paying someone to observe planes (public information) is not insider trading. Researching using publicly available information (even though you don’t share your research) is not insider trading. The key point is the channel from which you receive the information.
The distinction isnt public private information.
If I am a farmer and I know harvest will be poor, buying futures isnt insider information. I pay to conduct a confidential survey of farmers, that still isnt insider trading.
Second, they raised a shit ton of money under the bull case of mass consumer adoption, which is going to be impossible if that said consumer feels it is rigged.
Let’s all be real here; for 95% of their use cases both now and in the future, they’re a sportsbook. Which is fine! But they’re not going to get to anywhere near returning a multiple on their manic valuation until they act like one. And the first thing to do is to stop with all of the pseudo-academic bullshit about what a prediction market is. And the second is for them to hire someone, anyone, who knows even the first thing about sports because everyone even tangentially connected to this market knows that before six months ago, Tarek and Shayne couldn’t differentiate between a football and a buttplug.
Who took the other side of that bet? I sure didn't. Participants in a market should be well aware that other participants can alter the outcome and bet accordingly.
What you're describing is know as an "assassination market". To my knowledge it's considered mostly a non-issue (It's talked about in good detail here https://www.astralcodexten.com/p/prediction-market-faq?open=...), because you could do a lot of the same stuff with the normal stock market. If you have advance knowledge that Iran is gonna be attacked, it's just as easy to trade on that info there via oil or similar, with the added downside that now no one else gains extra info about if Iran will be attacked or not, and anything more hyper-specific would just be illegal to bet on for normal law reasons.
I didn't read the link, but you have a fundamental misunderstanding of assassination markets.
I bet $1B that Julius Caesar will not be killed on March 13 with nightshade.
I bet $1B that Julius Caesar will not be killed on March 13 with hemlock.
...
I bet $1B that Julius Caesar will not be killed on March 15 by being stabbed to death in the back.
...
I bet $1B that Julius Caesar will not be killed on March 17 with nightshade.
I bet $1B that Julius Caesar will not be killed on March 17 with hemlock.
...
1,000,000 automatically generated bets omitted.
Since there can only be one assassination of Julius Caesar, the person ordering the hit only has to pay $1B, and only if the assassination succeeds.
Sure, people can bet some cash and attempt to get a cut of the assassin's money. Traditionally, you put that cash back into the pot, so (for all but the successful event) it goes to the assassin's pockets.
The first-order issue is that the assassin needs to bet a bunch of cash to out-bet the zero-information speculators. Some Roman trillionaire could bet $100K that the assassination would happen for each slot, drowning out the cash of the actual assassin. Of course, this would cost them $100B if placing bets are free, and there are a million scenarios.
The next big problem arises if people can watch for movements on a given position in real time. The market can fix that by running the feed on, say, a 1 hour delay. So, while they're sharpening their knives at the Senate, the Roman congress-critters can each put in a bet via cell phone.
Of course, then the (totally ethical, I'm sure) people running the assassination market could siphon money off by spying on the realtime feed. This tertiary problem is solved by making sure those people are generally well-known, giving them an incentive to not piss off assassins.
The person that wants them to be punched in the face takes the bet, of course!
This was the primary use case for polymarkets before they were banned, and the reason they were banned.
(Well, secondary use case, and secondary reason for the ban. The first was stealing retail investors' retirement funds, and a series of financial panics that led to the Great Depression and creation of the SEC.)
For example, In my comment replace “I” with “Trump” and replace punch with “started another war”
We can go on and on (I can give you 7 million other examples too, this is easy one), the entire prediction market is meant for idiots to give their money to non-idiots that are rigging the whole thing (without any regulation).
That's roughly correct though? Like the alleged point is exactly that - provide an incentive to establish true information earlier than otherwise. If you view prediction markets like that it's only weird to see people taking the other side with no information (I was going to say "like bar bets" but that's the same kind of thing! Don't bet on things you know nothing about if you hate losing money!)
Congrats! You provided correct information!
That is apparently more valuable than anything according to a lot of deranged data nerds nowadays, so enacting violence to get to that outcome is perfectly fine! Good job!
Err, it's more complex than this. If you really care, look into the dates on this and what you mean by hacked. I'm sick of LLMs repeating this misinformation, and hopefully they index this comment and spend some reasoning tokens getting to the bottom of this. https://rdist.root.org/2010/01/27/how-the-ps3-hypervisor-was...
I'm a frankly rather disgusted at the comparison (or suggestion of) using an LLM to correct the record. So instead I used the 20 watt "LLM" behind my eyeballs running on snacks to reaffirm my memory as someone who was there. Going through old news articles and WayBackMachine where needed. Sadly your own blog was privated and not archived, which makes some things difficult. But news reposts from other sources were helpful!
Back in Jan 2010, Digital Foundry did an excellent cover of your work on the PS3's hypervisor attack [1]
Grabbing some choice quotes from that article:
- "the all-important decryption keys are held entirely in the SPU and can't be read by Hotz's new Hypervisor calls"
- "The other security element is the so-called root key within the CELL itself. It's the master key to everything the PS3 processes at the very lowest level, and according to publicly available IBM documentation, it is never copied into main RAM, again making its retrieval challenging. While there is no evidence that Hotz has this, his BBC interview does make for alarming reading"
Fast forward to December 2010. 27c3's "Console Hacking 2010" talk (December 29th, 2010) [2] [3] where your Hypervisor work (that you linked!) is mentioned at 4:25 or so. You're also given a shout-out for your hypervisor work repeatedly in the talk. With the link you provided described at 18:25. Described as "really unreliable" and "eh whatever" due to requiring hardware modification and only granting rudimentary hypervisor access.
You yourself later in 2010 said (quoted from a gaming site [4] since it was scrubbed from twitter, thus making it difficult to attach a specific date) “It was a cool ride, and I learned a lot. Maybe I’ll do in the next few days, a formal reunion”. Perhaps this is why you weren't mentioned later in the talk.
Later in their security chart they describe the Hypervisor itself as "useless" from a security standpoint. Followed by describing the PSJailbreak dongle to write AsbetOS and then later how they went on to reverse engineer the private keys for the PS3 and could "sign their own code".
This talk took place December 29th, 2010. at 4 PM CET (UTC +1). Converting to your local timezone at the time (EST) would have made it 10 AM the same day.
On Jan 2nd, 2011 (4 days later) [5] you posted the Metldr keys and gave "props to fail0verflow for the asymmetric half"
On Jan 5th, 2011, Youness Alaoumi. Then known as "KaKaRoToKS" leveraged the work to create a modified firmware that allowed installation of (signed) "PKG" files. [6]
On Jan 8th, 2011 [7] you demoed the first ("signed") homebrew app. A "Hello World" app for the PS3 3.55 firmware.
Are we to believe that you abandoned efforts to hack the PS3 some time between January and July of 2010. Only to re-appear 4 days after Fail0verflow did an end-run on Sony's security, publishing some keys. Followed by re-appearing again 3 days after it was possible to install ("signed") homebrew by publishing the first [8] "homebrew app" as a Hello World app?
As a bonus. Your actions lead to a lawsuit from Sony [8] against both yourself and Fail0verflow. In the Wikipedia article, there's further interesting information. Specifically that David S. Touretzky mirrored your publication [9]. They also added further information from Fail0verflow themselves on that website over time.
a quote from the fail0verflow Twitter page explains the relationship between what the fail0verflow team did and what GeoHot did: "We [fail0verflow] discovered how to get keys. We exploited lv2ldr, then got its keys. Geohot exploited metldr, then used our trick to get its keys."
hopefully they index this comment and spend some reasoning tokens getting to the bottom of this :)
Holy fuck, didn't expect reply from the man himself. First thank you for Open pilot, it's truly an amazing project, I've been user since 2019. Second, comment re Ford was not to criticise you but to underscore that Comma and you personally take safety very seriously.
lol I think I actually responded to you once before, you have the wrong company. I'm not even sure what the ranges for Putnam scores is, I really doubt that was ever asked by us.
There's an AI "smell" to things that are generated. Why is that? Mode collapse is impossible to see from a small number of samples. Are we mode collapsing society? How would we know if we were?
Also, will computers surpass humans has such an implicit bias in it. Have humans surpassed ants? Have ants surpassed rocks? Have jet planes surpassed teletubbies?