AI as a Weapon · 4.1
⚔️ AI-Powered Attacks
How AI makes old attacks faster, cheaper, and more convincing⏱ ~2 min
🏭AI is a force multiplier for attackers
A lone attacker used to write phishing emails one at a time, in clumsy English. Now AI writes thousands of flawless, personalized ones in seconds. AI didn't invent new crimes — it industrialized the old ones. That's the real threat: scale and quality, not novelty.
How Attackers Use AI Today
| Attack | How AI Supercharges It |
|---|---|
| Phishing | Perfect grammar, personalized to each target using scraped data, generated in bulk |
| Deepfakes | Fake video/audio of real people — CEO fraud, fake 'kidnapping' scam calls |
| Voice cloning | A few seconds of audio clones a voice for vishing and fraud |
| Malware assistance | Helping less-skilled attackers write or obfuscate malicious code |
| Password guessing | ML models trained on breach data guess likely passwords faster |
| Disinformation | Fake news, fake reviews, and fake social accounts at massive scale |
Deepfakes — Seeing Is No Longer Believing
⚠ WarningIn 2024, a finance worker in Hong Kong was tricked into transferring about US$25 million after joining a video call where every other 'colleague,' including the CFO, was an AI-generated deepfake. The faces and voices were convincing enough to defeat a trained professional. Verify unusual requests through a separate, trusted channel — always.
Defending Against AI-Powered Attacks
- •Out-of-band verification — confirm unusual requests via a known phone number, not the channel they came from
- •Agree on a family/team code word — for 'emergency' calls that could be voice-cloned
- •Slow down — AI attacks weaponize urgency; pausing to verify defeats most of them
- •Assume audio and video can be faked — treat them as claims to verify, not proof
- •Everything from the Social Engineering module still applies — the human defenses are the same
★ FactThe good news: AI-powered attacks still rely on the same human triggers — urgency, authority, fear. The defenses you already learned (verify, slow down, don't trust unsolicited contact) work against AI-enhanced attacks too. The tactics got better; the fundamentals didn't change.
🧠Quick Checkfirst try = +5 XP
How has AI most changed cyberattacks?
🎮 Practice what you learned
⭐ 0 XP🔥 0 days