"When AI Becomes the Hacker"
Bruce Schneier, an internationally renowned security technologist, has been examining the potential unintended consequences of Artificial Intelligence (AI) on society, particularly how AI systems could evolve in a way that enables them to automatically and inadvertently abuse societal systems. Schneier highlights the idea of the AI being the hacker rather than malicious actors hacking AI systems. The main question posed in his research is, "what if artificial intelligence systems could hack social, economic, and political systems at the computer scale, speed, and range such that humans couldn't detect it in time and suffered the consequences?" It's the stage at which AI systems can creatively find hacks. Schneier points out that AI systems are already doing that in software to find vulnerabilities in computer code but are not very proficient at it. However, AI systems will likely improve while humans stay the same in their capability to discover vulnerabilities. He predicts that AI systems will soon be able to defeat humans in capture-the-flag hacking contests because AI technology will evolve and surpass human abilities. We should be prepared for AI systems that can create their own solutions regarding hacks and vulnerabilities, and the use of this evolution by humans to make money. According to Schneier, the biggest risk posed by AI systems is that they will find a way to hack rules without humans realizing it. The core problem is that AI doesn't have the same human cognitive functions like empathy or a hunch that could allow it to know where not to cross the line. There are many studies on incorporating context, ethics, and values into AI programs, but they still aren't built-in functions of AI systems used today. Schneier admits that the idea of AI systems being hackers remains speculative, but it's an issue that should be considered and addressed. He recommends using AI to improve defense activities, such as finding and fixing all vulnerabilities in a program before it is released. This article continues to discuss key insights shared by Bruce Schneier in his research on the potential dangers of AI hackers.