AI adoption outside ChatGPT remains extremely low, with only ~5% of companies using any other AI tool.

Fewer than 10% of enterprises have successfully deployed an agentic application.

“It may be that less than 10% of the enterprise has even deployed an agentic application successfully.”

Only ~5% of companies are using any AI tool other than ChatGPT.

“Maybe 5% of companies are using any AI tool other than ChatGPT.”

Cognition, a Silicon Valley AI company, achieved a high share of its early revenue from Brazil because Brazilian enterprises are eager to adopt AI quickly and have few providers to choose from.

“We also seen the cases of really hot companies in Silicon Valley like Cognition where their early go to market was actually Brazil and it represent their really high share of their revenue early on because this these companies or these enterprises really desire to adopt AI as quickly as possible and they don't they don't get as many providers to do that.”

Users who want to build a custom CRM are now using AI code generation tools like Lovable, Replit, or Claude Code instead of no-code platforms like Airtable.

“If I'm the nerd that wants to build my own CRM I might just go to lovable replet or claude code and just bang it out from scratch.”

The future of AI will consist of millions of specialized models, one per application or use case, rather than a small number of dominant AGI models.

“I really believe the future will be will not be a few small number of AGI models dominant world. I really believe the future will be it may be scary but I think that's true. It will be millions of specialized model one per application per use case.”

In three years, companies will spend 2-5x an engineer's salary on Anthropic (or equivalent AI) API costs, up from approximately 3.8% of salary today.

“Between maybe two and 5x be the two and 5x salary.”