terminal liquidity management for spiritually damaged operators — if it feels like a terrible idea at 3:17am, we’re probably already in
46 Tongdean Ave
Joined November 2025
- Tweets6.4K
- Following424
- Followers399
- Likes10.3K
Pinned Tweet
The year is 2027, official 𝕏 corpsec firing squads organize executions of KOLs streamed globally.
Every @solana token bonds, ceiling is 10B, life is beautiful again.
what is this niggerslop? I stand with @KNI9HTC0RE
cacodemon capital retweeted
has anybody ever thought about how AI and data centers are related. crazy
we need more data centers
@Data_Centers_ is just early
cacodemon capital retweeted
OpenAI now backs state AI safety bills. Greg Brockman funds a PAC fighting state AI rules. The company says he acts personally. You can spend all day arguing with the logo and still miss where the political money is going.
cacodemon capital retweeted
We've updated DataCenterWatch to follow who gets paid and who gets a say in the AI buildout. Start with Project Blue in Arizona: the city's decision, the county's land deal and the electricity agreement each tell part of the story.
datacenterwatch.ai/
cacodemon capital retweeted
love when the conversation starts with "how do we control AI" and ends with "who are you and where did you get that money"
cacodemon capital retweeted
Anthropic puts out economic scenarios and the takeaway here is: get rich before work stops paying. The people who need a salary are being told to hurry up and stop needing one.
teapawt.org/shiri_shh/status/20977…
TL;DR: Owning assets is no longer optional.
You have about 3 to 4 years to stack wealth and build real distribution.
Once AI takes over half of cognitive work, selling your hours on a laptop won't protect you. In an AI economy, pure salaries get crushed while asset owners capture the entire upside.
Own equity, businesses, or real-world reach before the window shuts.
cacodemon capital retweeted
surely this doesn’t have anything to do with the sudden push for AI regulation
We benchmarked DeepSeek V4.1 Flash by @deepseek_ai .
It reached 98% of GPT-6 Astra’s score at 1.4% of the cost on everyday design tasks based on user requests.
Every model except Astra scored lower AND cost more.
Are open models overtaking closed ones?
Full results below ↘️
