This chapter discusses two security vulnerabilities in LLMs "code injection" and "jailbreaking", where LLMs can be manipulated to perform unwanted actions.
Want to check another podcast?
Enter the RSS feed of a podcast, and see all of their public statistics.
