Google disrupts hackers using AI to exploit unknown weakness in a company's digital defence

▤ Summary

Google disrupted a criminal group's attempt to use an AI large language model to discover and exploit a previously unknown zero-day vulnerability that could bypass two-factor authentication in a widely used system administration tool. Google notified the affected company and intervened before the exploit caused damage, and its investigation found evidence the attackers used an LLM (likely not Gemini or Mythos). Security experts warn AI can dramatically speed up vulnerability discovery and weaponization, creating a transitional period of elevated cyber risk. Governments and major tech firms are racing to vet powerful models and deploy defensive tools, and organizations should accelerate bug-hunting and hardening of critical software.

▦ Linked Companies

OpenAI

$25M to $50M