> Let me put it this way, the hedge fund no doubt hired qualified IT specialists who know much more than you and get paid much more than you to handle their security

I am one of those kinds of specialists and I get paid pretty well LOL

Thank you. Final questions -

What happens if you order 100 widgets with a pay -on-delivery agreement, pre-sell 100 widgets to other party, and those widgets don't ever get delivered by the time specified in the contract?*

Because that's what's going to happen in the case of these shorts, right?

*In this scenario the other buyer has plenty of stock and this was just a 'top-up' order for the warehouse. Because of the billions of widgets this company goes through they have a steady flow from multiple suppliers so they are not make or break counting on you.

Well you have your opinion. I think Griffen is Madoff 2.0 just more advanced.

No, I don't even work in the financial sector. I just have a good idea about how corporate accounting, purchasing, trading work.

I answered that, but I should have been a bit clearer.

The person who ordered the 100 but only received 10 takes a loss on the value of 90 widgets on their books. They will not receive the product and can write off the resulting loss. Additionally, they have the right to make a bankruptcy claim against the seller for the value of 90 widgets and may recover a portion of that value. Making this claim may or may not be worth it based on the time and effort required, the likelihood of being paid, and the value of the contract.

Everything else I mentioned is created specifically to hedge against the likelyhood of this happening.

Businesses like known costs, even if those costs are sub-optimal. That's why businesses might buy insurance for contract fulfillment on large orders like like I mentioned. If the value of that contract is $500 million and they take delivery over time, a business would much rather pay a 3rd party say $750k to insure the small percent chance of losing most of that value from non delivery and simply not take the risk of losing that 470 million in product, or whatever it is.

But the short and most basic answer to your question is that the buying business just takes a loss.

Let me put it this way, the hedge fund no doubt hired qualified IT specialists who know much more than you and get paid much more than you to handle their security (not saying this to attack you, just making my point)

Like the guy that played 1.5 years of college football then thinks they know better than Bill Belichek lol

Something like that. They just shouldn’t let it near the financial and transaction records or correspondence.

Yeah, I don't think they would let it touch their systems. They just buy the license and run it on an off-site server where it can do it's thing to write targeted blog posts and "news" articles.

Let me put it this way, would you hire a hedge fund manager to manage your network security operations, configure your firewalls, set up your intrusion detection systems, etc.?

>This type of account isn't odd, it's absolutely the norm.

Right I understand that, I'm just wondering what happenes in the reverse where you order 100 widgets and the supplier takes your money then goes bankrupt after only shipping 10.

You will still end up in bankruptcy held responsible by your creditors, right?

There is nothing illegal about shorting over 100% float. We could argue about whether it should be illegal (probably should) but there's nothing saying they couldn't do it right now. Option volume on many stocks also adds up to more than the sum of the shares they promise. Should that be illegal?

If anyone is to blame, it's probably Robinhood. They should not have disabled buying.

That being said, again they didn't do anything illegal. They're allowed to change their offerings at any time and for any reason, and people using them agreed to that when signing up for the service. If you want a broker that doesn't fuck you over, then you're going to be paying commissions.

I say all this as someone who was there for the whole GME debacle and pulled a significant amount of money out of RH as a result. There very likely is nothing illegal going on here. Why cheat when you can make money without cheating?

I suppose I would agree. It's just laughable coming from them, however.

You work for Citadel?

The purchaser would have a claim against the bankruptcy for the value of the widgets, but would likely take a haircut on the value.

This type of account isn't odd, it's absolutely the norm. Selling promises for future delivery is baked into our economy at a very basic level. Sure, companies going bankrupt happens - but it's part of the cost of doing business and the likelyhood of it happening is low. If the buyer is big enough to hold sway or the order is large enough, they might have specific language about cases of failure to deliver. They might even buy insurance on the delivery if it's something gigantic. Say McDonalds pre-purchasing 200 thousand tons of next season's potato harvest. I'd eat my hat if a delivery contract like that doesn't carry insurance.

Would it be crazy to think that the hedge fund who posted the best return in history with resources in the billions has qualified IT expertise who has considered and mitigated this risk?

It's better for them to sell the data than giving it away for free.

Better than complying

Firefox combined with a VPN. You can make it super private if you adjust some of the settings

Edit: Here's a guide on how to do that

So who owns the remaining widgets if you go bankrupt after the first 10?

I know nothing about managing a hedge fund. I know some things about having novel technology in networks where you also have sensitive data - mostly, that you don’t want to be the first one to do it.

Yes, but to ensure you have a model that’s behaving in that way, with standardized controls, you need to first established what those standardized controls are and then figure out some kind of auditing and certification framework for saying “this version of the tool works that way and is safe to use in an environment with sensitive information/regulated data.”

These organizations shouldn’t be trying to roll their own secure instance of ChatGPT (they wouldn’t even know where to start) and I bet they don’t want to.

If you localize the instance within the company, or more specifically, within the teams with access to that data already, and run different instances for those outside of that group, its less of a problem. The model being local, and only allowing input local should limit the risks, although if it is still scrapping current data, who knows, could be a risk poin

Fucking Citadel.

Armchair hedge fund manager lol

That’s not the problem, the issue is ChatGPT piping things from their network back to OpenAI.