The economics of software are changing.
Last night, as we launched new model by @AnthropicAI Fable on @emergentlabs , I built a personal sommelier.
Not because it was technically impossible before.
Because it wasn't economically sensible.
For decades, software had to be built for large markets to justify the cost of building and maintaining it.
But when software can plan, execute, test, and iterate over days rather than minutes, entirely new categories become viable. Software
A research lab.
A small business.
A team of ten.
Even a single person.
The most important AI applications may not be the ones that serve billions of people.
They may be the millions of highly personalized applications that were never worth building until now.
The long tail is finally worth building.
Built with Fable on @EmergentAI
Last night, as we launched new model by @AnthropicAI Fable on @emergentlabs , I built a personal sommelier.
Not because it was technically impossible before.
Because it wasn't economically sensible.
For decades, software had to be built for large markets to justify the cost of building and maintaining it.
But when software can plan, execute, test, and iterate over days rather than minutes, entirely new categories become viable. Software
A research lab.
A small business.
A team of ten.
Even a single person.
The most important AI applications may not be the ones that serve billions of people.
They may be the millions of highly personalized applications that were never worth building until now.
The long tail is finally worth building.
Built with Fable on @EmergentAI
0:14