Exartpolation (@_Exartpolation_) header
Exartpolation (@_Exartpolation_) profile picture
phone
Joined April 2023
Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)Photo by Exartpolation (@_Exartpolation_)
Tarromin ranarr and guam
11
Exartpolation retweeted
Anthropic has filed an S-1 registration statement with the SEC related to its proposed IPO. Read more: anthropic.com/news/s1-sec Legal disclaimer: A registration statement relating to these securities has been filed with the SEC but has not yet become effective. These securities may not be sold, nor may offers to buy be accepted, prior to the time the registration statement becomes effective. This post does not constitute an offer to sell or the solicitation of an offer to buy these securities, nor shall there be any sale of these securities in any state or jurisdiction in which such offer, solicitation or sale would be unlawful prior to registration or qualification under the securities laws of any such state or jurisdiction.
1,353
2,374
2,449
18,245
9,648,808
Exartpolation retweeted
The problem isn’t that AI is intelligent. It isn’t. The problem is that too many adults (but not all) are at once mechanistic, imitative, unimaginative, stupid, formulaic, influenceable, & gullible, arrogant beyond their skill set, & too expensive to maintain.
423
2,032
333
14,255
1,091,769
Exartpolation retweeted
# Black October For anyone who knows me even somewhat well, it's clear that I am a 99th percentile tech optimist. However, the vibes have shifted for me quite a bit in the last 18 hours, and I feel a moral imperative to share what I'm thinking and feeling with my circles so that everyone can understand, and hopefully take seriously, what's happening right now. I think October 2026 will have many chapters written about it in the history books. You know those scenes in financial movies like Margin Call where someone realizes that they're all fucked, and no one else sees it yet? And that person feels like they're crazy, and then suddenly everyone is pulling an all-nighter in a boardroom and freaking out? Yeah, that's kind of how this is feeling right now in the SF/Bay Area/AI circles. We're 48 hours into what is being called Black October. In the last 2 days there have been several dozen cyber attacks which, in isolation, would each have been the worst to happen in an entire calendar year. Every sector from government to education to big tech and banks is being massively affected, and people are starting to freak out. In a nutshell, what's happened is that frontier AI models have gotten to a point where they are as advanced as the most sophisticated programmers, with one critical difference: they can work 24/7, and you can spin up 10,000 of them with a single command and they'll work together tirelessly until they accomplish their task. The most advanced models today are from companies like Anthropic and OpenAI, whose current best technology is so good that they haven't been able to roll it out to everyone yet because of the national security risks it poses. Meanwhile, the best open source models (where anyone can just download the model weights and do whatever they want with them, uncensored) are slightly behind the frontier but still 90-95% as capable. These models, developed primarily in China, are being used by black hat hackers to wreak havoc on the Western capitalist structure. Presumably, this has been part of China's plan the entire time, and we're seeing its implications fall out in real time. In red-pilled AI circles, we've been sharing and reading papers called things like "Situational Awareness" and "AI 2027" for the last year or so, which at the time felt a bit like near-term speculative fiction. They laid out the case for why Artificial General Intelligence (AI smarter than any human at any knowledge task) and Recursive Self-Improvement (models that are able to use themselves to get better) weren't half-baked future ideas anymore, but instead an inevitable outcome of the linear progress that's happening. These papers would then talk about things like "takeoff scenarios," trying to envision how that moment would come to be and what it would be like when it happened. It would either be a "slow takeoff" that would play out over the course of months and quarters, happening subtly and only obvious to a few. Or, it would be a "fast takeoff" where it would pretty much happen instantly, like a bit flipping from 0 to 1. From everything that I can see and am feeling, we are currently experiencing the fast takeoff scenario. Beyond the crazy security vulnerabilities that have come out recently, there have also been massive breakthroughs in life sciences, fundamental computer science, materials science, and pure mathematics that have all landed in the same time frame. While I still firmly believe that AI will be a net positive for humanity, it's more clear to me than ever that the societal and individual risk is very real, and is playing out NOW. For as much conviction as I have that this is happening, the thing that's caused me to lose sleep is that I don't really know what to do about it, or more importantly, what to tell you, reader, to do about it. So, the best I can do is make a few suggestions that have always been a good idea but are more timely now than ever: Set up 2-factor authentication for everything. I use Duo. Use a password manager across your devices. I use 1Password. Get an identity protection app set up. I use Alta. Move your IoT devices to their own network. Claude helped me do this. And last, but certainly not least, if you've been avoiding leaning into AI, see this as a catalyst to wake up and start using it. Don't just use the chat mode version of ChatGPT/Claude; click that Code/Work button and see how easily you can have AI write code to help you with your job. One of the biggest vulnerabilities was in a massively popular programming language called PHP, and the prompt that set it off was literally, "Find vulnerabilities in php." This is an unprecedented time for humanity, but as we've done with many things before, we will prevail so long as we are kind, loving, and open-minded to the world. Much love.
716
2,009
385
18,012
3,808,678
Human mote against ai = bank accounts and government IDs
16
Exartpolation retweeted
Replying to @vikktorrrre
AI, build me a magical theme park that not only rivals Disney World, but surpasses it in magical feels. Create an entire library of tv and movies with thousands and thousands of characters to tie intellectual property into the rides and experiences. I'm going to go have a beer.
1
643
Exartpolation retweeted
GameStop's bid for eBay: A Textbook Activist Play?
49
83
14
524
133,723
Bitcoin addresses are public 🤣🤣🤣
45
Exartpolation retweeted
Michael Saylor diluting $MSTR in order to buy bitcoin and then selling the Bitcoin to buy $STRC so that he can buy back $MSTR so he can sell more $MSTR to buy $STRC so that he can buy Bitcoin.
217
270
72
2,801
454,496
Exartpolation retweeted
Just wrapped our earnings call. It was a very strong close to what was a record fiscal year for Microsoft. · Annual revenue: $331B, +18% · MS Cloud: $214B, + 27% · And Azure: $100B, +41% And even bigger opportunity ahead! I wanted to share some more perspective on two areas of focus for us:
757
1,104
201
16,849
2,276,091
Exartpolation retweeted
I wrote about why we believe the future is for everyone. More coming about a positive vision for a world with superintelligence soon.
4,008
2,177
1,141
31,066
6,510,693
Exartpolation retweeted
I think the next moat is energy efficiency. $NBIS just published its first Sustainability Report! it delivers ~20% more AI compute per watt, runs at 1.25 PUE and uses 25x less water than the US data center average..
14
30
3
321
21,891
Exartpolation retweeted
Replying to @kevinxu
Microsoft earnings report. Also the "but ROI on capex spend tho" is such a mid curve take everyone's realizing. For those that pander this bear take, ask them to explain how big tech companies make money and how the Ai tech works before they get to numbers.
1
1
123
Crypto really lost the plot. Bitcoin was supposed to decentralize. Memes are too retarded. Perps still scarred from October 10 but gamblers gonna gamble
Attention world! We are very proud to officially launch the $JESSE vault, now live @pantheonvaults @jessepollak already has over 58k $JESSE holders. Functionally the coin has existed as a temporary attention trade. Starting today it becomes an attention economy. Now every builder on @Base has a transparent way to get their token into the hands of one of the ecosystem’s largest onchain communities. For projects: Deposit tokens into the Jesse Vault and they’re distributed over time to $JESSE stakers. For holders: Stake your $JESSE once and receive a stream of tokens from builders choosing to participate every single month! Projects get discovery and distribution to onchain verified investors without spending a dollar in stable coins. Holders have a reason to hold and get exposure to the latest and greatest builders shaping the Base ecosystem. Now the $JESSE token becomes infinitely more useful and valuable, and as holders stake to the vault it becomes more and more scarce. Simple and powerful. Thanks to the support of: @taniatare @bigwil @mizzysworld @SpikeCollects @WayfindersGG @0xPatchara @DadsGoneCrypto @CryptoGrit the Jesse vault is already receiving its first deposits for holders. If you’re a $jesse holder, simply start here: launch.pantheonvaults.com/va… If your a project looking to get in front of Jesse and his community either despair directly via the app or contact us for support. This is the turning point for creator economies, a huge step towards a more transparent onchain crypto economy and the beginning of a new and improved asset class that makes crypto great again!🏛️
1
108
Exartpolation retweeted
New interview! Why Bitcoin Miners are Powering the AI Data Center Boom w/ $MARA CEO Fred Thiel YT Link: youtube.com/watch?v=qOxGs8Zs…
2
9
30
6,877
Also they were the reason alt szn didn't happen
1
46
Perp dex meta killed crypto. No trader will trade with those liquidation wicks on chain
51
Exartpolation retweeted
TLDR: During recent volatility, Hyperliquid had 100% uptime with zero bad debt. This was Hyperliquid’s first cross-margin ADL in more than 2 years of operation. ADL does not change the outcome for any liquidated users. While some specific ADL providing trades were unfavorable, the aggregate effect of ADL was that traders realized significant pnl by closing positions at favorable prices that were only briefly available. -- It’s sad to see some people attack Hyperliquid to deflect from their own platforms’ issues. Solvency and uptime are the two most important properties of a financial system. These are table stakes for any trading system, and gaslighting to convince users otherwise is unethical and irresponsible. Below is more analysis on how Hyperliquid’s margining system handled the extreme volatility. Background on liquidations For a perps system to be solvent, every position must be backed by a minimum amount of collateral. This is called the “maintenance margin.” When positions do not meet the maintenance margin requirement, they are taken over by the system to be liquidated. Earlier today, many altcoins dropped by more than 50% in a short period of time. When this happens, long positions at 2x or higher leverage must be liquidated, or else the system accrues bad debt. There were billions of dollars worth of positions liquidated on Hyperliquid in a matter of minutes. In a permissionless system, each user chooses their own position sizing and collateralization. Any system that does not liquidate the necessary users is irresponsibly gambling with other users’ funds. On Hyperliquid, every order, trade, and liquidation is transparently verifiable onchain. Many other venues significantly under-report liquidation data. This cannot be compared apples-to-apples against the fully onchain picture of Hyperliquid. Background on HLP HLP is a protocol vault with permissionless deposits that 1) provides order book liquidity and 2) performs backstop liquidations. The first role is negligible, with HLP trading less than 1% market share. The focus of this post is liquidations. Liquidations are first attempted against the order book, and any user can provide liquidity to these market liquidations. Backstop liquidations occur when the order book does not have enough liquidity to absorb an undercollateralized position. In this case, HLP takes over the position along with its collateral. For improved risk management, HLP is split into several child vaults, and each liquidation is only sent to one child vault. Background on ADL Auto-deleveraging (ADL) is the liquidation mechanism of last resort, when market and backstop liquidations do not work. See Doug’s thread (link in reply) for a thorough explanation on the details of ADL. Every ADL event has two sides: the “triggered” side is undercollateralized, while the “providing” side is decided as a function of profitability and leverage used. Similar to backstop liquidations, even though providers to ADL are profitable on average, there are no guarantees for any specific event. Some ADL providing trades were unfavorable, such as when only some components of long/short portfolio were closed. The system is designed to minimize ADLs because they are unpredictable even if ADL providing trades are profitable on average. Because HLP is a non-toxic backstop liquidator, ADL is a rare settlement of last resort. As far as I know, this was the first cross-margin ADL event on Hyperliquid mainnet (ADL is more common for isolated-only assets such as hyperps, which are not backstop liquidated by HLP). Summary of events Over the course of 20 minutes, HLP backstop liquidated billions of dollars worth of positions. HLP's philosophy has always been to provide liquidity of last resort. Contrary to misconceptions, HLP is a non-toxic liquidator that does not pick profitable liquidations. Instead HLP is a public good for maintaining system solvency. In particular, Hyperliquid has no liquidation fees. HLP’s design, including its multi-component child vault system, is the product of countless simulations, and allows HLP to maximally serve the benefit of the protocol while managing its own risk. In fact, the liquidator child vaults of HLP themselves became undercollateralized in the course of backstop liquidating as many user positions as possible. This is by design, where child vaults are isolated from the other components of the overall strategy. HLP is treated no differently from other users when participating in ADL. In aggregate, HLP's child vaults were the largest addresses on the triggered side of ADL by more than an order of magnitude. The addresses on the providing side of ADL against HLP’s child vaults realized hundreds of millions of dollars in additional profit relative to the prices shortly before and after the dislocation. On other venues, the liquidation engine is not transparent and therefore may not be subject to the same strict margin requirements as for normal users. On these venues, the exchange could have backstop liquidated more positions, bearing increased solvency risk to extract hundreds of millions in business revenue. This is not an acceptable tradeoff for Hyperliquid. Finally, I know that this is a difficult time for many traders, and I hope the community can continue to support each other and grow together. As a contributor to Hyperliquid, I’ll continue to work my hardest to build the best possible platform that can house all of finance. Times like this highlight the importance of transparency and fairness in the financial system.
806
908
285
5,280
1,214,834
Exartpolation retweeted
*Warning* (VERY STRONG), Retweet this to let the Universe send you money, Huge Amounts of money very quickly. 888.
334
3,647
163
5,229
641,743
Exartpolation retweeted
Replying to @blknoiz06
it is bro
9
7
3
100
7,858