For most of its short life, Anthropic was known less for what it sold than for what it feared. The company built its name on caution, publishing research on how advanced models might mislead, manipulate, or slip beyond the control of the people meant to run them. In 2026 that picture has flipped. Anthropic is now one of the fastest growing businesses in technology, and the rise has turned it into something it never set out to be, a target.
The revenue story
The numbers behind the change are hard to ignore. Anthropic's annual revenue, measured as a run rate, has climbed at a pace few software companies have ever matched, moving from a few hundred million dollars to several billion in little more than a year. Most of that money comes not from consumers chatting with a friendly assistant but from businesses wiring Claude, the company's family of models, straight into their own products and operations.
Built on code
The clearest engine of growth is software development. Claude has become the model many programmers reach for first, valued for writing, fixing, and explaining code with unusual reliability. Young companies that sell coding tools have built much of their worth on top of Anthropic's systems, and large enterprises increasingly route their developers through the same models. That concentration is a strength and a weakness at the same time. It gives Anthropic a loyal, high spending base, and it leaves the company exposed if a rival ships something cheaper or simply better.
Why success invites attack
Commercial momentum on this scale rarely goes unanswered. OpenAI, Google, and a well funded crowd of challengers are chasing the same enterprise budgets, and they have noticed where Anthropic earns its keep. Price cuts, lookalike products, and louder marketing now point straight at the coding market Anthropic helped define. Being the leader means everyone else measures themselves against you, and prices against you too.
The fight in Washington
The other target on Anthropic's back is political. Dario Amodei, the company's chief executive, has spent the past two years warning loudly about the dangers of powerful AI, from mass job losses to misuse by hostile states. Those warnings have earned attention, yet they have also placed Anthropic at odds with parts of a government that wants the technology built quickly and built at home. The company argues for guardrails while insisting it should be among the few allowed to build the most capable systems. Critics see a tension in that, a firm asking for rules that would conveniently leave it standing.
A culture under the microscope
Inside the company, the seriousness that shaped its research now shapes its image. Admirers describe a mission driven team that treats safety as a duty rather than a slogan. Detractors reach for a harsher word, calling it almost a cult, sure of its own virtue and certain that the future of the technology should pass through its offices. Either way, the intensity that helped Anthropic recruit talent and raise money has become part of why it draws fire.
The cost of being out front
Rapid growth carries its own bills. Training and running frontier models demands enormous amounts of computing power, and the contracts that secure it run into many billions of dollars. To cover them Anthropic keeps raising money at valuations that would have looked absurd only a short time ago. The math holds as long as revenue keeps climbing and customers keep paying premium prices. If competition flattens those prices, the gap between what Anthropic spends and what it earns could widen fast.
What to watch next
The question for the rest of 2026 is whether Anthropic can stay both principled and dominant. Its product lead is real, its customers are sticky, and its brand still carries weight with engineers and policymakers alike. Leadership in this market, though, is a moving target, and the company has picked a harder path than most by trying to win commercially while preaching restraint. Success made Anthropic a target. Keeping the position will be the harder act.
