AI agents automate vulnerability research, end attention scarcity

The collapse of "attention scarcity" means unglamorous but critical targets like medical devices and legacy infrastructure are now at extreme risk from automated exploitation.

• –LLMs act as universal jigsaw solvers, using stochastic search and latent knowledge of bug classes to discover and verify exploits
• –Open-source maintainers will be swamped by a torrent of reproducible, high-severity vulnerability reports they lack bandwidth to patch
• –AI agents render "security through obscurity" obsolete by easily translating assembly and decompiled code
• –The impending wave of AI-driven exploits could trigger reactionary regulation, potentially criminalizing essential vulnerability research