Cracking Down on Cyberthreats in the AI Era with AWS's Clayton Smith

Solving security problems for customers of cloud giant Amazon Web Services (AWS) is Principal Security Specialist Clayton Smith's primary mission. But sometimes it can feel like playing cops and robbers, with attackers and defenders both racing to put generative AI to work first.

AWS has invested heavily in AI-powered security innovation, including systems like Mithra, which maps relationships across billions of internet domains to spot malicious infrastructure before it's used in an attack, and MadPot, a network of AI-enhanced honeypots that studies attacker behavior in real time. Clayton sees these systems as proof that AI-powered defense can now operate at a scale no human security team could match on its own.

But the technology cuts both ways. The same natural language tools that make security more accessible to non-specialists also lower the bar for attackers to craft convincing phishing attempts, generate malicious code, or automate reconnaissance. Clayton's take is that this isn't a reason to slow down. It's a reason enterprises need to move at least as fast as the threat, with an acceptable risk profile they've actually thought through rather than inherited by default.

One underappreciated effect of AI in this space: it's democratizing cybersecurity careers. Natural language security tools mean an analyst doesn't need a decade of scripting experience to ask a system the right question and get a real answer. That's good news for an industry that has struggled for years to fill open security roles.

As threats continue to evolve, so must the professionals defending against them. The best strategy, Clayton said, is to “lean into that change, but still question everything.”

“Take a step back, take a deep breath, and embrace the fact that this is happening,” Clayton said. “As security professionals, how do we make sure it’s the best, most secure experience for our customers and our organization?”

Learn more about Backstory’s approach to AI and data security.