Welp, that happened faster than I predicted. Thought it would be end of 2027, then early 2027, but agentic traffic growing so fast that bots have now passed human traffic online for the first time in the Internet's history. https://t.co/2zX5bHdhsa
Since #TokenOptimization is the all the rage, I thought I'd take a crack at characterizing the market based on what I've seen so far. I cribbed the OSI model to create the #AiSI Model: 7 Layer approach to manage AI costs from GPU up through the application (roughtly).
Come on, Claude, get it together. You're a hot mess today. Why would you take a process that was working seamlessly and just switch it up and do it in a way that fails miserably?
*rubs temples*
@stevehou I think it’s probably $500 annual run-rate. And probably “all-in” costs including tokens BYOM (build your own model) and consultants. I could easily see an F500 at that level.
Finally watching #OKCvSAN game 6 (I know I’m late), but if the @NBA is smart, they’d separate these two and make them meet in the #NBAFinals. They’re going to be great for a while.
When you see Founder Mode from a non-founder
When you find a VP, an exec, that actually cares as much as a founder, or almost as much …
Promote them fast. ASAP. Double their equity.
And … never let them go.
Big, proprietary Foundational Models - YES!
Tuned & customized models for enterprises - YES!
Inference & AI Infra providers - YES!
Big 3 CSPs - YES!
On-device inference - YES!
I guess it is human nature to think zero-sum (SAD!), but sure feels like there's demand for all of it.
@dhaber Hey, I tried to get an elephant to dance. But there’s a fundamental tension between the “sexiness” of venture & reality of a risk accepting function.