KevinSimback

辣评：Anthropic 的定价是为了达到绝对的完美
> Mythos 必须交付
> 企业必须在代币支出上保持松弛
> 采用率必须在全年保持一月至五月的节奏
> 计算能力必须扩展
> 不会遇到政府或社会层面的阻力
我不赌他们到年底前能达到 $100b 的运营水平，但他们不能有任何失误

This latest cohort of crypto AI projects is quite different
People still using token price as an indicator of product quality
But after digging deep on this cohort of projects I can say this relationship doesn’t hold
The alpha lives in really evaluating the products

My top focus area within AI right now is finding companies that do either of the following:
> help reduce frontier lab API bills
> provide / enable alternatives to frontier lab usage (the 90% of tasks that don’t need it)
From pure tokenmaxxing to token cost optimizing

Using LLM subscription plans today feels like going to an extravagant all-you-can-eat buffet
It costs you nothing to try everything so you consume for the sake of it
Enjoy while it lasts, next step is token dieting, then come the token equivalents of GLP-1s so we consume less

This is a bit off topic, but retatrutide (small dose 1x/wk) + 10-20g/day creatine feels like a cheat code
Down a few lbs, strength up 10-15% over 2 months, and feel refreshed with about an hour less sleep

It is pretty clear to me that the best opportunity for business executives and owners to use AI is as follows:
> discover the top performers across your workflows
> build agents based on these top performer’s workflows
> put those agents into self-improving loops on those workflows
> let those self-improving agents coordinate across workflows to reinvent how the business is done
> humans to oversee all this
The tools are all there, it’s just a question of which orgs can stomach the idea

We will look back at 2026 as the golden age of subsidized intelligence
For the same price as Netflix, you can get near unlimited access to frontier models
For $200/month you can employ a swarm of agents to build whatever you want
And enterprises are getting sweetheart deals to ensure provider lock-in
All subsidized by investors
But the tides are beginning to turn
Frontier labs are raising prices (modestly, for now) and beginning to move advanced capabilities behind usage based pricing models
The closer they get to post-IPO, the more they have to price for value, not adoption
What does this mea

The AI pie is growing so fast it’s easy to miss trends within the trend
One worth watching:
Serious products are stacking proprietary RL on open-weight bases instead of just routing to frontier APIs
They’re bringing post-training in house and getting better results than just routing to frontier APIs
Cursor’s Composer 2 is the best example
25% of compute on the Kimi base and 75% on Cursor’s own RL using their coding trajectories
Many others are doing this quietly, many more will go this route
The base model is the commodity and the post-training is the moat
The only complication is that the po

While everyone is paying attention to the frontier labs, there is a market segment that:
> wants open weights
> can't or won't use Chinese models
> values vendor neutrality
I'll give you 1 guess as to who fits this perfectly
The question is how big and fast this segment grows