Jump to content

Talk:Bankruptcy of FTX

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Requested move 16 November 2022

[edit]
The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion.

The result of the move request was: Moved to Bankruptcy of FTX. Consensus to move to 'Bankruptcy of FTX' as it is less ambigious and more accurate. (closed by non-admin page mover) – robertsky (talk) 03:08, 24 November 2022 (UTC)[reply]

FTX financial crisis → Collapse of FTX – "Crisis" doesn't seem like it's the best title for what's happening. It's concise and it's descriptive, but it's not the ideal title. I'm proposing the title Collapse of FTX due to it being able to outline the entire saga of FTX, including the bankruptcy, while alternatively, as suggested by @MainlyTwelve:, Bankruptcy of FTX would be a proposal I would also be inclined to support. Either way, a more descriptive title than "crisis" works better for the article at large. InvadingInvader (userpage, talk) 22:57, 16 November 2022 (UTC)[reply]

  • Support to "Bankruptcy of FTX". I think it's a little more encyclopedic. I've added the existing article content to that page and can undo my edit removing it to restore it if we're agreed. — Mainly 23:03, 16 November 2022 (UTC)[reply]
Support to "Bankruptcy of FTX" since it would match with Bankruptcy of Lehman Brothers Vctrbarbieri (talk) 18:00, 22 November 2022 (UTC)[reply]


The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Help expanding section on effective altruism

[edit]

I've started a subsection about the FTX collapse's impact on the EA movement: § Impact on effective altruism. Please help spruce it up by adding links to reliable sources with commentary on EA's role in the scandal. This is a particularly balanced, detailed one that covers a lot of what I've added about the debate over EA's consequentialism and risk-taking. Qzekrom (she/her • talk) 06:55, 22 November 2022 (UTC)[reply]

Page title again

[edit]
speaking as a true geek of yore

Back in the 1990s I read Bruce Schneier's Applied Cryptography cover to cover more than once. I even interviewed with Zero Knowledge Systems to work alongside Ian Goldberg. I tend to turn interviews around and I started to ask some hard questions about the ethical boundaries of total anonymity. The person interviewing me got a look on his face that I was too much of a perceptive thinker, but in the wrong box. They were looking for someone less heads up and more heads down for that particular role. His concluding line was "well, you certainly asked a lot of hard questions", but it was also kind of dismissive. I think more than kind of dismissive, as in, I might already have been viewed as actively counterproductive.

It was, however, my natural wheelhouse. Onion routing was considered something of a subtle concept at the time. But for me, I was like, how is there any other possible way to do this? Ideally you also did this exchanging constant bandwidth between all pairs of nodes, with all the traffic appearing uniformly pseudorandom to anyone without the keys. Ideally the fill traffic was generated at the edges and handled exactly the same as payload traffic internal to the onion, except for having a lower queue priority. This helps mitigate timing attacks which are nominally "air gapped", but potentially visible as ripples in power supply. And payload packets should never have a forwarding probability greater than say 10% node-to-node, so that a lightly used onion doesn't end up with deterministic forwarding patterns. Latency is your friend.

When Bitcoin came along, I read enough to grasp that it was a nice piece of work, but then I lost interest. Most of the elements were fairly obvious, but like the iPhone (also obvious at the component level), the perfect integration of all those moving pieces is a work of art. "Ian, is that you?" I wondered briefly. I don't think I saw his name mentioned too often as the man behind the curtain, but one could make a worse guess. Especially since I had originally suspected that Zero Knowledge was just the first step toward a larger plan whose exact nature was obscure to me, but most definitely right out of 'Cryptonomicon.

Stephenson also includes a precise description of (and even Perl script for) the Pontifex cipher, a cryptographic algorithm developed by Bruce Schneier for use with a deck of playing cards, as part of the plot. The Perl script was written by cryptographer and cypherpunk Ian Goldberg.

Of course, Cryptonomicon arrived years later, but not a single thing inside it surprised me, so far as the mathematical adventure went. That ethos was entirely inside the room in the early days of Zero K.

This entire space reminds me of a dollar auction which reminds me of Thomas Pynchon's Gravity's Rainbow. You can live in that world and not lose your shirt, but the order of business is wall-to-wall paranoia. There's a neurological principle that what fires together wires together. After your neurons fire together long enough, you have basically become a mirror of your environment. Paranoia is insidious that way. In the long run, it's a cognitive root canal. At least for me, it was, because I felt I had bigger fish to fry. I wanted to know how the entire human project fit together, and every waking moment was going to be devoted to that inquiry. As a deep thinker, your final gear in what you bring to the table is that you become one with your cognitive environment, via the fire-together principle.

Because cryptocurrency is inherently unhinged, meaning that your fear of dollar auction never entirely goes to zero, there's no way to side on the sidelines without a substantive cognitive trip down the rabbit hole of what every other batshit person in the ecology believes they believe. Idiots disguise themselves a clever people. Clever people disguise themselves as idiots. And the smart money runs their reputation through an onion router of smart-stupid-smart-stupid-smart-stupid under the total number of layers exceeds our short term working memory (which is actually less than seven according to fMRI studies).

Sviatoslav Richter once confided to a friend, after hearing Glenn Gould perform in Russia: "you know, I could play as well as Gould, but I would have to work so hard"". Yeah, I had the mental tools to mostly figure out the number of onion layers on each individual's reputation stack, but I would have had to work so hard. Somehow Gould did this without noticing how hard he was working. But Richter would have noticed how hard he was working every damn minute. I was merely Richter level, not Gould level. Genius-level fools rush into civilization-scale dollar auctions where 2nd rate angels fear to tread. I was clearly a 2nd rate angel, so I took a hard pass.

Briefly, I had a college roommate who was probably a step up from Ian. I won't name names, but his was a strange path, and we finally lost touch. PhD at Stanford in real time emergency medical expert systems. Not the easy project: What's the best order to ask questions to arrive at the correct diagnosis with the least amortized effort?, but instead, the slightly harder problem of ordering that enquiry while the patient is bleeding out right there in front of your eyes.

That soon got him recruited by a quantitative trading house in New York. So I asked my college friend, a multiple Putnam fellow, how do you stack up in the room? "Merely average. At best."

Renaissance Technologies, is that you? There can't have been too many rooms out there where my old pal could have been anywhere remotely close to merely average. It was either RT, or some dark boiler room fully staffed up to stalk a gorilla. "Wow. Well how good are the best people?" I don't remember his precise answer, but it was along the lines of "I can barely hold their jockstraps". Exactly right. When you're next door to Alan Turing, his doorbell seems 10 feet off the ground. You might be the second smartest man in one of the five smartest rooms on one side of a global conflagration, but you still need a step stool to ring Turing's doorbell. Postmodernism has decreed that IQ is passe. That sure the fuck hasn't been my own lived experience, but what the heck do I know?

the actual matter at hand

Now, to the matter at hand. From my perspective, having arrived at the scene of the crime after several decades of studious avoidance, the current page title Bankruptcy of FTX does not cut the mustard.

The good aspect of this title is that it is 100% objectively true. Always a good starting point in fast-moving story with many moving parts and conflicting counternarratives.

Now for the bad parts. Too many people don't know the difference between bankruptcy—often a loose synonym for Chapter 11—and true insolvency. Bankruptcy is supposed to be a mechanism to stop the train before a cash flow crisis explodes into true insolvency. None of the creditors want this, but they can't sanely negotiate any intermediate solution while the train continues to chug along under suspect management.

So you have a B$30 valuation, and you're walking a thin plank between a Ponzi scheme on one side and a dollar auction on the other side, and maybe the plank is entirely stout and legal, your performance of walking this plank along the straight and narrow is of paramount importance. Perception management is the entire ball of wax, unless you really can halt a run by conjuring up instant liquidity.

I read dozens of accounts last night, and every single version of the story notes that a potential conflict of interest existed between FTX and Alameda Research. Tedious, onerous, multiple-layers-of-red-tape financial controls at this boundary should have been a gold-plated belt backed up with platinum suspenders. But most accounts paint a picture of the formal controls at this interface having been much closer to The Wolf of Wall Street after the protagonist obtains a rare and dangerously expired bottle of Lemmon 'ludes.

Okay, fine, but the true nature of this apparent self-dealing remains to be unequivocally established. But at least the phrase 'hit by force majeure' should not be passing anyone's lips, unless you're the kind of sloppy thinker who would describe the final state of the Wolf's Lamborghini in the same terms. That amounts to debating whether a wine contains a hint of stone fruit until the plummy tobacco.

So the big red emergency stop button is mashed and the train comes to a halt, and a new crop of sober adults enters the room. How bad is it? Let's consult the ledgers, the records, and the controls.

Ten minutes later: Uh, wait just a second, where's all the money?

It's been a long time since I read the book about LTCM, but what I recall is that they had over 60,000 positions on the books when they halted the train, and barely a dime out of place. Their back room team that put the software together to accomplish this feat was snapped up instantly elsewhere in the ecosystem. The same team at FTX probably wound up in a fistfight to monopolize the one bottle of liquid paper, so that they could circulate their resumes at all.

There's a pretty high degree of certainty already that the books were a shambles, if not outright fraudulent. None of the direct communication around this issue by the principle suspects has been anything to write home about. The state of affairs surrounding this system rises to the level of scandal even if not fraudulent, and I would say even if the negligence is merely incompetent. At a B$32 valuation in this space, on a tightrope of self-interest stretched between the missing towers of the World Trade Center, there's simply no get-out-of-jail-free card on your central records degenerating into a Puzzle Palace. Total, instantaneous disqualification from any hint of a pretext of someday purchasing Goldman Sachs (that kind of loose talk was actually recorded).

I also don't like "collapse" because LTCM collapsed, because of hubris on nearly the grandest scale ever recorded, but with their house almost entirely in order. And in fact, but for liquidity, their final position was likely solvent, since most of their trades were guaranteed arbitrage bets, given enough time. Had their risk portfolio remained sufficiently decorrelated, there never would have been a problem, but then the fall of Russia was a rug-pull unconceived in their greedy philsophy.

Lehman Brothers is not a good model, either. One sarcastic view of Lehman is that Lehman collapsed instead of Morgan Stanley (in much the same boat) because more of Lehman's creditors were overseas, and more of Morgan's creditors were domestic. By overseas, I mean Asian. "Screw the slopes" might have been the quiet voice in that nasty bit of business. It's pretty much the most plausible hypothesis after blind caprice. For optics, we need to sacrifice at least one goat, ASAP; which goat will generate the least proximal blowback? Clearly, the one that leaves fewer powerful American interests holding an empty bag.

As you can probably tell, there's at least one corner of my brain that runs to pedantic, and the title that works best from that corner of my inner world is this one: FTX fiscal control saga.

I use 'saga' because it's already a scandal that we don't yet know that this doesn't likely rise to the level of culpable fraud (a central harmonic of 'scandal' in a page title) among the cast of characters whose primary day job was to know better—but very much an uncertain and unfinished scandal. A central figure who ought to know has already declared the kemptness of their books as "worse than Enron".

Mark to market, wrapped inside a mystery, inside an enigma, served up steaming hot in a slippery sauce of slovenly audit trails.

I don't think my proposed title will generate any traction, but I did want to put myself on the record that the current page title distracts the focus of the article very far away from the central circus ring. This is one of the prices Wikipedia sometimes pays for hewing as much as possible to the straight and narrow (as FTX surely didn't).

But we shouldn't be happy about this state of affairs, and we should bail on the current page title at the nearest sensible exit, lest our prudence continue to distort history in a realm where distortion is too often an entire mode of existence.

That my first and final two cents. Please pardon my borderline neurodivergence for running toward wall of text ... — MaxEnt 21:55, 24 November 2022 (UTC)[reply]

Can't use 'miasma' or 'contretemps' in a page title (why shouldn't a cryptocurrency end up with a bad cybersmell?), but I did come up with FTX fiduciary blemish and bankruptcy as a passable halfway house. — MaxEnt 23:22, 24 November 2022 (UTC)[reply]
If/when fraud becomes better proven perhaps we could change the article title to something concise that captures what happened, like "FTX scandal". Your two suggestions, while appreciated, are, I think, too baroque. — Mainly 17:22, 25 November 2022 (UTC)[reply]
Maximum Entropy, I fully understand the scope of the accounting and controls horror at FTX. The former Enron guy who was called in to be the "new" CEO expressed this in no uncertain terms. Since you mentioned the Necronomicon... it could even be described as a Lovecraftian horror by analogy. Quoting NPR:

"Sam what have you done?," tweeted Sean Ryan Evans, host of the cryptocurrency podcast Bankless, after the bankruptcy filing..."This is the direct result of a rogue actor breaking every single basic rule of fiscal responsibility"... The ultimate impact of FTX's bankruptcy is uncertain, but its failure will likely result in the destruction of billions of dollars of wealth.

Via Reuters, FTX's Bankman-Fried begged for a rescue even as he revealed huge holes in firm's books (16 Nov 2022), "Some of the $10 billion in removed customers' money went to cover losses that Alameda sustained earlier this year on a series of bailouts, including in failed crypto lender Voyager Digital. To conceal the transfers of customer funds to Alameda, [FTX executive Gary] Wang, a former Google software developer, built a backdoor in FTX's bookkeeping software, the people said. Bankman-Fried often told employees tasked with monitoring the company's financials that the book-keeping system was "the ultimate source of truth" about the company's accounts... But the backdoor, known only to his most trusted lieutenants, allowed Alameda to withdraw crypto deposits without triggering internal red flags. Alameda accumulated a huge holding of FTT, valued at around $6 billion before last week... It used the FTT reserves to secure corporate loans... This meant that Bankman-Fried's business empire was dependent on the token."
NB Despite all of this, Sam Bankman-Fried is NOT on the lam, in hiding, with the authorities on his tail. He was scheduled as a keynote speaker at the New York Times Dealbook Summit event on November 30th, and it seems that he will still be there. I would leave the article title as is for now.--FeralOink (talk) 18:24, 25 November 2022 (UTC)[reply]
That's way too ambiguous and long of a title. Bankruptcy of FTX is a good title, and while not my personal preference (I kind of like Collapse of FTX more), it does the job better than any other proposal. InvadingInvader (userpage, talk) 18:42, 28 November 2022 (UTC)[reply]
Returning to the subject, we might need an article along the lines of "2022 Cryptocurrency exchange crisis", as the problems spread beyond FTX but seem to be confined to cryptocurrency exchanges. Latest bankruptcy, BlockFi[1] Is there such an article yet? John Nagle (talk) 17:28, 29 November 2022 (UTC)[reply]

Current Event Tag

[edit]

Hi, I have added the currenct event tag to thiss article. There seems to be new information breaking pretty frequently about what went on within FTX, what the bankruptcy will look like etc, so it seems fair to note that information may still change rapidly/sporadicaly.

MJD (talk) — Preceding undated comment added 13:15, 1 December 2022 (UTC)[reply]

"Liquidity crisis"

[edit]

Some editors want to describe the events leading up to the bankruptcy as a "liquidity crisis." However, that term was pretty much only used by RSs in the first few weeks after the bankruptcy, before people knew what had actually happened.

FTX was not a bank, and could not have suffered a liquidity crisis the way that banks usually do. Customer deposits were not supposed to be loaned out, unless customers explicitly opted in to the margin lending program, and even those that didn't opt out lost their money. Can Sun, a former FTX lawyer, testified in Bankman-Fried's trial that customer deposits were supposed to be kept safe and be available for withdrawal at any time. See https://www.ft.com/content/df9aa325-6a3b-453d-a718-8380e4a3b6d7.

Using the term is confusing and obscures what actually happened. I think the most appropriate way to describe the collapse is to say that a spike in withdrawals revealed that customer deposits had been misappropriated (as has been proven in court). — Preceding unsigned comment added by RaspberryCandy (talk • contribs) 00:10, 5 November 2023 (UTC)[reply]

@RaspberryCandy But the question is: What caused the spike in withdrawals? Also, on WP, we do not rely on the most recent sources to describe events. We instead use the most reliable ones, regardless of the date. Quetstar (talk) 02:16, 5 November 2023 (UTC)[reply]
The current wording seems accurate, but I also see why it seems misleading. Saying this was caused by the 'liquidity crisis' seems a bit euphemistic, since it was more properly caused by fraud. Even without this particular liquidity crisis, this would've still been fraud, and moving this term down from the second sentence might help with an already extremely complicated and confusing topic.
As for reliability, age is a factor, especially for rapidly-changing topics, such as breaking news stories. Out-of-date sources are inherently less reliable. We're not hide-bound to include out-of-date information merely because it was presumed to be accurate at the time. Grayfell (talk) 02:48, 5 November 2023 (UTC)[reply]
Yeah, I agree. Quetstar (talk) 06:27, 5 November 2023 (UTC)[reply]
There have been dozens of NYT articles about FTX-related events since the collapse, none of which are less reliable than older reports. None of the recent articles describe what happened as a "liquidity crisis." I don't know why we'd have to be chained to a misleading, confusing and euphemistic description that was primarily used when no one knew what was going on.
See Wikipedia:AGE MATTERS. "Sometimes sources are too new to use, such as with breaking news (where later reports might be more accurate)." This seems to apply here.
It's easy to find up-to-date sources. See, for example, this recent NYT article describing the FTX collapse. It doesn't use the word "liquidity" anywhere, but rather explains that the spike in withdrawals was caused by growing public concern and panic about FTX after CoinDesk leaked an Alameda balance sheet and Changpeng Zhao responded by announcing he'd sell his FTT. RaspberryCandy (talk) 04:44, 5 November 2023 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:Bankruptcy_of_FTX&oldid=1205704253"