Inside Cyber Warfare, 3rd Edition

Chapter 7. AI

Machines take me by surprise with great frequency. This is largely because I do not do sufficient calculation to decide what to expect them to do.

Alan Turing¹

The pursuit of artificial intelligence (AI) is almost as old as computing. Stanford and MIT’s AI labs were both founded by the same man, computer scientist John McCarthy. McCarthy arrived at MIT in 1956 as a research fellow and opened the AI Lab with Marvin Minsky in 1959. In 1962, he moved to Stanford and founded the Stanford Artificial Intelligence Lab. 

In 1956, McCarthy expressed their major challenge in a way that was similar to the test that computer scientist Alan Turing proposed in 1950: “The artificial intelligence problem is taken to be that of making a machine behave in ways that would be called intelligent if a human were so behaving.”²

More than 60 years later, computer scientists working in AI still anticipate that a sentient AI with human-level intelligence will emerge from its silicon base to deliver a techno-utopia to the human race, if it doesn’t kill us first. 

This chapter will explore present harms and future risks associated with the rapid adoption of AI, especially as it impacts national security. I’ll detail the most pressing present risks and the harms that have resulted from them, explore future risks, and provide recommendations for prevention and mitigation. 

Defining Terms

In order to navigate the world of AI, it’s important to have some agreed upon definitions of terms. It’s ...