Anthropic at $965 Billion: The Startup Taking On Big Tech Over AI Safety

Dario and Daniela Amodei founded Anthropic in 2021 with an idea that seemed almost heretical at the time: an AI company that puts safety and model alignment at the absolute center, not as an afterthought. While Silicon Valley was screaming "move fast and break things," they chose the opposite path. The result? A valuation of nearly a trillion dollars and the respect of people who actually know what they're talking about.

The real pressure comes from Pentagon battles and geopolitical tensions. Anthropic doesn't want its models to become propaganda weapons or military control tools. It's a risky position in a world where the U.S. government is pouring billions into military AI and competitors don't lose sleep over ethics. But the founders explained why this matters: while others play poker with humanity's future, they're at least trying to put some rules on the table.

The paradox is fascinating. Anthropic claims to put safety first, yet it lives in an ecosystem where speed wins. How do they balance it? Simple: they convince investors that "responsible" doesn't mean "slow." Their Claude models work well, they're reliable, and you don't have the feeling tomorrow morning they'll declare war on you. It's a value proposition that, thankfully, is actually working.

This Bloomberg interview is rare because the founders rarely open up like this. They talk about the company's origins—two researchers who worked at OpenAI said "nope, I don't like this approach" and built something different. They also discuss long-term vision: it's not about winning the race for the biggest models, but building AI that humans can actually control and understand. Sounds idealistic? Yes. But the valuation suggests the market is starting to believe pragmatism and ideals don't have to be enemies.