Gili Raanan of Cyberstarts: AI will double all of human technology within 25 years — and that's the threat

Summary

Gili Raanan, founder of Cyberstarts, frames cybersecurity as an existential acceleration problem, not a tactical one. Human technology will double within 25 years, a span no living person has witnessed. The last doubling took roughly 170 years from the development of writing to the steam engine. The next doubling arrives in 25 years. In 100 years, humanity reaches 2000% technological capability, and people will regard today's technology the way we regard cave drawings.

The threat landscape has become asymmetric. Offensive actors and state sponsors can adopt AI tools immediately and apply them to attacks. Defensive adoption lags severely inside large enterprises. Bad actors gain access to frontier AI capabilities faster than good actors, creating a genuine risk if that gap persists.

Frontier labs like OpenAI, Anthropic, and DeepMind monitor their own platforms and can detect when customers spike activity on malicious prompts. Traditional cybersecurity vendors including Palo Alto Networks, CrowdStrike, and Wiz maintain access to frontier models and can fight with bigger weapons than attackers. The equilibrium assumes continuous innovation, with new companies like Onyx Security and Self AI entering to fill gaps.

On whether labs themselves threaten incumbent cybersecurity businesses, Raanan applies a classic pattern. Short-term disruption is oversold, long-term impact is undersold. Over the next two years, frontier labs' focus remains elsewhere, so competitive threat is low. By five to seven years, labs could dominate. Three competing cohorts will emerge: traditional cybersecurity platforms, cloud providers, and AI platforms. A year ago, he would not have named the third group at all.

Raanan's recent predictions have already missed by underestimating speed. Twelve months ago, he predicted a $1M ARR single-founder company within two years; Base reached $3.5M ARR before acquisition in under a year. He predicted a $100M revenue company with fewer than 100 employees in two years; Cursor hit $100M ARR with fewer than 20 employees.

Fortune 100 companies will soon operate cybersecurity functions with fewer than 10 employees, down from thousands today. AI will transform cybersecurity from a profession into a function. The pressure is real, the direction clear, and the timeline compressed.