What is the danger?
Attackers use AI to write flawless phishing messages, find software vulnerabilities, write malicious code and — most recently — run large parts of an attack with autonomous AI agents that act with little human direction.
How it happens
- Personalised phishing in perfect language, at massive scale.
- AI that searches for and exploits software vulnerabilities.
- Autonomous agents running thousands of attack steps on their own.
- Attacks on AI systems themselves: prompt injection, data poisoning.
Real cases
Anthropic reported disrupting what it described as a largely AI-orchestrated cyber-espionage campaign that used its Claude model against dozens of targets (2025).
Source: Anthropic ↗Google researchers said a cybercrime group likely used AI to build a “zero-day” exploit — a hacking tool for a flaw nobody knew about (2026).
Source: Bloomberg ↗Hackers used autonomous AI agents in attacks on Taiwanese government agencies, which experts called a possible first fully autonomous attack of its kind (2026).
Source: CNN ↗
Warning signs
- Unusually well-written urgent emails asking you to log in or pay.
- Messages that reference real details about you or your work.
What you can do
- Turn on two-factor authentication everywhere — ideally with an app or security key.
- Keep software and phones updated.
- Never log in through a link in a message; open the site yourself.
Frequently asked questions
What is the danger of cyber in AI?
Attackers use AI to write flawless phishing messages, find software vulnerabilities, write malicious code and — most recently — run large parts of an attack with autonomous AI agents that act with little human direction.
How does it happen?
Personalised phishing in perfect language, at massive scale. AI that searches for and exploits software vulnerabilities. Autonomous agents running thousands of attack steps on their own. Attacks on AI systems themselves: prompt injection, data poisoning.
How can I protect myself?
Turn on two-factor authentication everywhere — ideally with an app or security key. Keep software and phones updated. Never log in through a link in a message; open the site yourself.
Latest updates on cyber
Hackers used autonomous AI agents to attack Taiwan
Attackers deployed an autonomous AI system against Taiwanese government agencies, in what experts believe may be the first known fully autonomous cyberattack on government targets — raising fears about the future of cyberwarfare.
Google: criminals likely used AI to build a zero-day exploit
Google's threat researchers said a cybercrime group likely used AI to create a hacking tool for a previously unknown flaw in widely used system-administration software — the first such case the team has caught.
What do you think?
Seen this danger up close? Worried about it? Share it with the community.
Share your thoughts on cyber