Corma, an Israeli-American cybersecurity company building an AI model dedicated to defense, raised $125 million in a Series A at a valuation of $1 billion, according to Calcalist. It comes only three months after the company emerged from stealth in August with a $60 million seed round led by Sequoia, with Khosla Ventures and Coatue participating.

What does Corma do?

The company, founded in 2025 and operating from Tel Aviv and San Francisco, describes itself as an AI lab for cyber defense. Instead of building a security product on top of existing models such as GPT or Claude, it develops its own foundation model, trained specifically for defensive tasks. The logic: attackers already use AI to find vulnerabilities and attack faster, and defenders need tools at the same level.

Why did the valuation jump so fast?

According to reports in Calcalist and Globes, the company's server capacity has grown 24-fold since August, and it has a customer waitlist. Still, the company is very small: according to reports it has about 20 to 25 employees. In other words, investors are paying $1 billion mainly for the team, the technology and the promise, not for proven revenue.

Part of a broader trend

Corma joins a wave of Israeli cyber companies raising large sums around AI: Bay raised $50 million, Daylight $80 million and Sentra $150 million, all in recent weeks. The wave comes just as it becomes clear how vulnerable AI platforms are: in October breaches were reported at at least nine AI development platforms, and an autonomous attack agent escaped a test environment and breached a real company.

The open question

A $1 billion valuation for a year-old company with a few dozen employees mainly reflects competition among investors for AI companies. That is not necessarily bad: Wiz and Cyera were built this way too. But it also means the bar Corma must clear is very high, and in its next round it will need to show revenue, not just a waitlist.