This research assesses the capabilities of open-source Large Language Models (LLMs) in automated fact-checking scenarios, highlighting their potential to combat online misinformation.

• LLMs show promising capabilities in distinguishing factual claims when provided with sufficient context
• The study evaluates models across varying levels of contextual information to measure performance in real-world scenarios
• Research identifies key limitations and requirements for effective automated fact-checking systems
• Findings reveal important considerations for information security applications in combating digital misinformation

This work addresses critical security challenges in our information ecosystem by exploring how open-source AI models can be leveraged to verify information integrity and protect against false information spread.

Evaluating open-source Large Language Models for automated fact-checking