Sunday 19 July 2026 18:03:05 GMT+02:00
Netcrook
#HalluSquatting
HalluSquatting Shows How AI Assistants Can Be Tricked Into Pulling the Wrong Code
Researchers demonstrated a naming attack against AI assistants that can move from hallucinated lookups to remote code execution and, in some cases, malware delivery.
When a Model’s Guess Becomes a Supply-Chain Trap
A newly named technique turns AI coding mistakes into a security boundary problem, showing how a hallucinated identifier can become a dangerous trust decision.
Hallucinated Names, Real Control: The AI Weak Spot Behind a New Botnet Playbook
A reported technique called HalluSquatting shows how an LLM’s confidence can become an attacker’s entry point when agents are allowed to fetch or run what the model invents.