Automated vulnerability detection existed long before the era of artificial intelligence; many people have been doing this for a long time. All the existing tools and documentation paved the way for what AI is capable of today. Furthermore, undisclosed vulnerabilities still do exist—often discovered by state-sponsored actors who exploit them for opaque purposes (frequently operating outside the legal framework of the country in question).