The archive · AI & Models · Product decision · 2023–2026
Exaforce bets AI agents stop cyberattacks in real time; $125M Series B at $725M
Security teams drown in alerts, so Exaforce's Exabots triage and stop threats live; $200M raised, 20 customers since launch.
Exaforce
What the business is
Exaforce sells an AI-native security operations platform whose Exabot agents triage alerts, investigate threats and act on them in real time, with a natural-language 'vibe hunting' feature that lets teams test attack hypotheses.
Starting capital：$200M total: a $75M Series A in 2025 plus the $125M Series B announced 2026-05-12.
How it started
Exaforce was founded around 2023 by Ankur Singla with a deliberately simple mandate: apply AI to catch and stop threats as they happen. The company spent two years testing its platform with design partners before bringing the product to market in the fourth quarter of 2025, betting that security teams were drowning not in real attacks but in the false-positive alerts that bury the real ones.
What happened
After a $75M Series A in 2025, Exaforce announced on 2026-05-12 a $125M Series B valuing the company at $725M, with HarbourVest, Peak XV, Mayfield, Khosla Ventures and Seligman Ventures participating. Its Exabot agents use deep data analysis to automate security operations, a claim backed by a promised reduction of up to 90% in manual work, and the recently introduced vibe hunting feature lets analysts query the platform in natural language, for example asking whether the company saw new attacks from a specific country. Since the Q4 2025 launch Exaforce has added 20 customers including Replit and Guardant Health, and says high-profile attacks have removed the need to sell the problem, with customers now asking how to operationalize the product.
No ending yet — it is still running.
Background
Security operations centers face a flood of alerts, most of them false positives, so real threats hide in the noise. Exaforce, founded around 2023 by Ankur Singla, bet that AI agents rather than more human analysts are the answer. Its Exabots use deep data analysis to automate triage, investigation and response, with the company claiming the platform can cut manual, time-consuming security tasks by as much as 90%.
The product came to market in the fourth quarter of 2025 after two years of testing with design partners. A distinguishing feature is vibe hunting, which lets analysts interrogate the platform in natural language, such as asking whether the company received any new attacks from a particular country, turning a hunch into an investigation. The approach found early buyers fast: within months Exaforce had 20 customers, including Replit and Guardant Health, and expected to reach 40-50 by the end of 2026.
Capital followed the momentum. After a $75M Series A in 2025, Exaforce raised a $125M Series B announced on 2026-05-12, valuing the three-year-old startup at $725M, with HarbourVest, Peak XV, Mayfield, Khosla Ventures and Seligman Ventures participating and total funding reaching $200M. The company competes with AI-security startups 7AI, Dropzone AI and Prophet Security, as well as the SOC platforms of Palo Alto Networks and CrowdStrike.
What has to be true
- The pain was real and measurable: analysts get hundreds of alerts a day and cannot tell which ones matter, so automation that triages for them has immediate value.
- The mandate was narrow and executable: catch and stop threats as they happen, rather than bolting AI onto an existing toolchain.
- Two years of design-partner testing meant the Q4 2025 launch shipped on real workflows, and the vibe hunting feature made live investigation approachable for analysts.
- Market timing did the selling: high-profile AI-driven attacks removed the need to convince customers the problem existed.
- The risk is competition: agentic security operations is crowded, and winning against both startups and Palo Alto Networks and CrowdStrike requires the claimed 90% labor reduction to hold up.
What can be applied
A security startup can win by automating the tedious 90%: once attacks made the need obvious, customers stopped asking why and started asking how to operationalize it.
Aftermath
As of 2026-09-05 Exaforce is scaling: $200M raised, 20 customers since the Q4 2025 launch and a stated target of 40-50 by end of 2026, competing against 7AI, Dropzone AI, Prophet Security and the SOC platforms of Palo Alto Networks and CrowdStrike.
Sources
- Exaforce raises $125M Series B to build AI for catching and stopping cyberattacks as they happen
- AI cybersecurity startup Exaforce raises $125 million
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card