@haussofballoons
Account based inUnited Kingdom
About this account
- Account based in
- United Kingdom
- Connected via
- United Kingdom App Store
Account-level information from X, not a live location or the device used for a specific post.
cross the chasm
Joined August 2022
- Tweets1.5K
- Following441
- Followers482
- Likes2.9K
Exhibit 2:
Love to see it.
solana:Grass7B4RdKfBCjTKgSqnXkqjwiGvQyFbuSCUJr3XXjs
sm retweeted
nice to see, but for all paying attention keep in mind that these numbers do not yet reflect 1) pretraining data purchase/budget cycles at the major labs (which are much more back half of the year heavy) and 2) the 2-3 order magnitude larger TAM in live retrieval + inference time search products (that this team will inevitably dominate in)
Grass DataCo Ltd., a wholly owned subsidiary of Grass Foundation, provides data infrastructure to leading AI labs training frontier models.
We're publishing an independent attestation of its revenue through Q2 2026, conducted by Regen Financial.
Full report: grass.io/resources/grass-att…
You’re telling me what @0xdrej has been publicly shouting at the risk of his entire career & criminal proceedings has been true all along!?
Shock horror!
🍄
ok, read this to understand SHROOM
on RH RWAs are all a lil volatile, and are being paired with memes that are super volatile, while all the pools for them have high fees
its only natural and expected for a new medium of exchange to develop to pair with ALL the RWAs to eat up all the fees and get routed through regularly
I think people don't understand this, because they're likely thinking 'but we have stablecoins you cretin'
The thing, if the medium of exchange is volatile, there will be frequent arbitrage opportunities against the RWAs for MMs and aggregators to inevitably route through to make a few bps.
The bigger the depth of liquidity is on those arb opportunities, the more money those pools will make for the new volatile medium of exchange token/protocol
the game is to grow liquidity enough in pairs with that new MOE that it then begets more volume, that then begets more fees, that then creates more liquidity through fees being used to shure up the pools (and to give back to holders via distributions of RWAs/buybacks).
it's a flywheel (shoot me, but it is)
SHROOM is the category leader.
It has nothing special about it in terms of tech moat, other than it has that je ne sais quoi to become the breakout token for this purpose, and a cracked thoughtful dev team who are tweaking things to max out the fee capture/liquidity expansion over time.
buying SHROOM is a bet on it being the new robinhood 'liquidity layer' token, and it having a pareto distribution (capture 80% of the market whilst there's a longtail of smaller tokens trying the same thing) by being the first mover to reach escape velocity.
If the RWA/meme or RWA/volatile-token trading volume continues on RH (seems likely to me), SHROOM could become very, very, very big.
dyor blah blah
I asked the painting robot to paint Citrini and it painted a lemon. Well, it tried to paint lines.
So, if the will-smith-eating-spaghetti curve holds here this robot arm will be painting the Sistine chapel in about 18 months.
sm retweeted
Guys, the world is so compute constrained. I honestly don’t understand how anyone can think otherwise.
Almost nobody is seriously using AI today, and we’re already hitting capacity. Everything is a bottleneck: land for data centers, permits, electricity, grid connections, chips, construction, cooling.
Demand is growing far faster than supply, and expanding supply is insanely expensive.
In this game, the best-positioned companies are the ones with massive net income from non-AI products. They can fund hundreds of billions in CapEx while everyone else has to raise money just to stay in the race… and remember if you are not at the frontier you have no pricing power so you won’t be able to fund your compute need.
Kimi K3 has received far more love than we expected, and our GPUs are feeling it.
Over the past 48 hours, demand has pushed close to the limits of our current capacity. To protect the experience of existing subscribers, we're temporarily pausing new subscriptions and prioritizing compute for current members. Existing subscribed users are not affected.
We're adding capacity as fast as we can and will reopen new subscription spots in batches.
Going forward, we'll also split membership into two more focused plans: Kimi Membership for Kimi Web, App, and Work; and Kimi Code Membership for coding workflows. This will help us match compute more precisely and keep the experience stable.
Thank you for your patience and understanding!
I'm using 5.6 Sol to find a new tumble dryer and this mfer pulling up arxiv papers as part of its analysis
The data & scraping infra $GRASS has built get more valuable over time
AI Labs continue to require higher quality & tighter curation in their datasets, which is only going to become more important given the outsized impact data quality has on model performance
YC partner leaving to join Anthropic’s compute team because
“as we enter the early stages of recursive self-improvement, availability of compute becomes one of the most important issues to solve.”
We need more compute.
$GLXY
Personal update: I'm taking a leave of absence from YC to join Anthropic. I'll be working with @NotTomBrown on the compute team.
Powerful AI has the potential to improve the life of every human on earth and, as we enter the early stages of recursive self-improvement, availability of compute becomes one of the most important issues to solve.
I'm excited to get started 🚀
not as glamorous but you can 2x your fable 5 limit by using it in cowork.
attach your local repo and code away.
I guess $SPCX is now a frontier AI lab too
Grok firmly in the race
Incredible
“Grok 4.5 is efficient and fast. It uses half as many tokens per task, delivers higher tokens per second, and costs less than half as much as comparable models, at $2/M input and $6/M output tokens.”
base:0x9b5e262cf9bb04869ab40b19af91d2dc85761722 bulls help me out please, here are my assumptions on first pass. What have I misunderstood? What am I missing?
Copy & pasted thoughts:
>i cant work out what nock is actually trying to be
> i can see how they can bootstrap proofs via merge mining; economically it makes massive sense for any gpu network to also post proofs to nock for any inference bcoz they then get paid twice
> that proof then helps secure the chain, but then what?
> supply is capped & logan has talked about becoming hard money. basically comped bitcoin mining to nock mining; each btc block mined is mining for the sake of mining; each nock block mined is a proxy for useful work which ye fine, i buy that thesis
> i dont see how the proofs become something valuable thereafter and all merge mined issuance then doubles up as sell pressure
> where it is potentially valuable is when a customer specifically wants verifiable inference which is a much rarer case and potentially $NEAR TEE attestation competes for that
> logan also spoken about the ability to spin up these networks for any type of useful work - if they're leaning into this hard money narrative then its potentially v exciting;
> whitepaper talks about private payments, bid auctions, compliance checks as potential demand side for verified inference
> but zero evidence of demand side showing up yet which i get - they're bootstrapping the supply side