all InfoSec news
Meta's Llama 2 LLM is still prone to hallucinations and other severe security vulnerabilities
April 17, 2024, 9:19 p.m. | Cal Jeffrey
TechSpot www.techspot.com
Meta's Llama LLM performed poorly in a recent third-party evaluation by AI security firm DeepKeep. Researchers tested the model in 13 risk-assessment categories, but it only managed to pass in four. The severity of its performance was particularly evident in the categories of hallucinations, prompt injection, and PII/data leakage, where...
Read Entire Article
ai security assessment evaluation hallucinations injection llama llama 2 llm managed meta party pass performance prompt prompt injection researchers risk security severity third third-party vulnerabilities
More from www.techspot.com / TechSpot
This genius tool ensures flawless thermal paste application every time
3 days, 9 hours ago |
www.techspot.com
Jobs in InfoSec / Cybersecurity
SOC 2 Manager, Audit and Certification
@ Deloitte | US and CA Multiple Locations
Emergency Management Invoice Compliance Reviewer
@ AC Disaster Consulting | Denver, Colorado, United States - Remote
Threat Intelligence Librarian
@ Microsoft | Cheltenham, Gloucestershire, United Kingdom
Cyber Content Operations Manager - Remote in UK
@ Immersive Labs | United Kingdom
(Junior) Security Engineer (m/w/d)
@ CHECK24 | Berlin, Germany
Cyber Security
@ Necurity Solutions | Bengaluru, Karnataka, India