Unreal…
Elon: Don’t use Starlink to kill people.
Zelensky: I used Starlink to kill people.
Legacy Media: Elon is a monster for stopping Zelensky from using Starlink to kill people.
Democrats: We must investigate Elon for stopping Zelensky from using Starlink to kill people.
The way to think about this is:
How many super smart and dedicated people do you want to hire? And they keep getting better.
Zooming out, GPUs are still renting ahead of a standard 6-year depreciation schedule
Rental price vs. schedule:
- H200: +143%
- B200: +111%
- RTX PRO 6000: +73%
- H100: +54%
- A100: +9%
Most of the gap opened up after April, when token demand took off
Charts of the Week: a16z.news/p/charts-of-the-we…
bitslurp retweeted
I’m telling you, Wall Street is short the right tail of compute.
They cry that there won’t be demand.
We won’t have enough supply!
Stay long the right tail.
bitslurp retweeted
It's very very different, this time.
AI Hyperscalers Net Debt/EBITDA at 0.1x vs. Internet Backbone Providers at 3.9x in 2000, a 39x difference.
The dot-com companies were funding fiber rollout with debt and capital market issuance.
Today's hyperscalers are funding AI capex internally from operating cash flow.
bitslurp retweeted
no I get it
we should be freaking out
a mere
[checks notes]
400% year over year growth at the frontier
“Some AI companies are doing phones and are telling Qualcomm we need to be able to run 100 Billion parameter models on the phone by 2028.” -Qualcomm CEO
Elon: “Problem is RAM supply.”
Joe Blow on X: “ $MU is cyclical and overvalued at FPE of 7”
Solana Securitize Samsung
S tier
BREAKING: Securitize launches Securitize Stocks on Solana.
U.S. equities backed 1:1 by real shares on @Securitize's registered broker dealer platform, with applicable voting rights and dividends for eligible investors in the U.S., E.U. and other permitted jurisdictions.
xitter.gq/Securitize/status/2108…
bitslurp retweeted
The compute needed for the stage of AI we’re about to enter is going to be insane.
Personal agents, agent swarms defending enterprises, agents that review all of our code for security issues, agents that process nearly all enterprise data in workflows, background agents working on tasks 24/7, and on and on.
This will take a mix of inference volume that is going to keep growing by orders of magnitude, as well as infrastructure that agents need alongside the tokens. They need computers, networking, file systems, and so on.
We’re only in the early stages of what this buildout is going to look like.
.@auchenberg calculated that Muse needs 65,000 CPUs and 75PB of DRAM to serve 100 million users. That's $2.8B in infra for one app.
It made me think. Facebook alone has over 3 billion users. That's nearly 300x scale for one platform. And they have more platforms with similar numbers.
Just do the math. There aren't enough CPUs in the world to run all of this. A worrisome constraint we're heading toward.
Correct perspective.
This is incredible. Wow!
In a certain very particular sense, this is now the world’s top mathematician.
Congratulations @sama! So exciting!!
bitslurp retweeted
The Titan Analytics Alpenglow dashboard has been expanded to Testnet. Congrats on a successful migration! @solana
titananalytics.io/analytics/…
There has been some serious dashboard performance improvements too. Check it out.
bitslurp retweeted
Replying to @yagiznizipli
get rid of the bug which shows me a tantalizing tweet for 1/50th of a second when I open the app only for it to be lost forever
bitslurp retweeted
the problem with these AI labs is they don't hire enough actual engineers to tell the researchers to shut the fuck up



























