On the 12th of May, Google Cloud’s GTIG team intercepted an AI-built zero-day before it was deployed, and the headlines called it a sign of things to come.
But that’s the thing: AI-powered attacks aren’t new.
In September 2025 Anthropic admitted Claude drove 80 to 90 percent of operations in state-based cyber-attacks.
In February 2024, Microsoft and OpenAI were reporting AI-based LLM attacks.
CrowdStrike says attacks are up 89 percent year-on-year.
The innovators in cybercrime are growing into AI just like you are, and the old-school gangs are catching up fast because they know where the value is.
If you’re waiting for the AI cyber-attack wave to arrive, look around, because you’re already standing in it.
#cybersecurity #zeroday #AI
But that’s the thing: AI-powered attacks aren’t new.
In September 2025 Anthropic admitted Claude drove 80 to 90 percent of operations in state-based cyber-attacks.
In February 2024, Microsoft and OpenAI were reporting AI-based LLM attacks.
CrowdStrike says attacks are up 89 percent year-on-year.
The innovators in cybercrime are growing into AI just like you are, and the old-school gangs are catching up fast because they know where the value is.
If you’re waiting for the AI cyber-attack wave to arrive, look around, because you’re already standing in it.
#cybersecurity #zeroday #AI
- Category
- Cybersecurity
- Tags
- #cybersecurity #AI #zeroday

Comments