Experts warn malicious AI skills are hitting more victims than ever — with one family amassing 1.7 million downloads
Source: TechRadar· Sead Fadilpašić· August 7, 2026
SynaBot summary
Malicious AI applications are increasingly targeting users, with one family of rogue skills reportedly downloaded over 1.7 million times. These compromised tools are designed to steal sensitive user credentials, posing a significant security risk.
Key takeaways
- Rogue AI skills are actively stealing user passwords.
- Millions of downloads indicate widespread vulnerability.
- Security researchers are identifying numerous dangerous variants.
- Vigilance is crucial when selecting AI tools.
Why it matters
Users of AI assistants must be vigilant about the third-party tools they integrate. The proliferation of malicious skills highlights the need for careful vetting of AI applications to prevent data breaches and protect personal information.
This story was reported by TechRadar. Read the full original article:
Read on TechRadar 
