The old answer was a seed round and a year of runway. The new answer is a rounding error on your credit card and a few weekends of your attention. But the honest breakdown is more interesting than the headline — because what's actually expensive about a startup in 2026 is not what it used to be.

She arrived in Canada as a refugee at eight, edited the Harvard Crimson, and managed billions at a sovereign wealth fund before she ever wrote a line of code. Then she bet a struggling translation startup's entire future on transformers — twice — and built one of enterprise AI's most quietly formidable companies. This is how May Habib turned a near-death pivot into a $1.9 billion full-stack AI platform, and what her refusal to be a wrapper taught the industry.

At twenty, Aidan Gomez was an intern at Google Brain — and one of the eight authors of the paper that quietly rewired the entire field of AI. Instead of chasing consumer chatbots or preaching about superintelligence, he built Cohere into a multibillion-dollar company selling enterprise AI to banks, hospitals, and governments. His is a story about betting on a different path, and being early enough to define it.

Bret Taylor co-created Google Maps, sold two companies, ran Salesforce as co-CEO, chaired Twitter's board through the Musk takeover, and now chairs OpenAI's board. He could have done anything next. He chose to start over — building Sierra, a company betting that AI agents are the new atomic unit of software. Here's how one of tech's most decorated operators became a founder again.

He co-invented the technique that half the AI industry now runs on, then left the lab to prove it could be done properly. From a childhood in Amsterdam to Facebook AI Research to a company built on making language models tell the truth, Douwe Kiela's path is a study in conviction. This is how a researcher turned one paper into an enterprise AI company — and why he insists the unglamorous problems are the ones that matter most.