Imagine a world in which artificial intelligence has gone rogue—the robots have revolted against their masters and have now enslaved all of humanity. There’s no more natural beauty in the world and everything is awful.

Get that out of your system? Good.

The reality of malicious AI, at least in the near future, is far less dystopian. However, it is a reality, and it’ll be here sooner than people might think. As with all disruptive technology, after blissful early adoption soon follows abuse. And if cybercriminals know one thing, it’s how to profit off a trend.

Therefore Malwarebytes Labs’ latest report takes a pragmatic approach to evaluating the dangers of AI and machine learning (ML), looking at exactly how these technologies are being used today, their benefits, and our concerns for near-future abuse. Why? So that developers, security professionals, and other organizations can incorporate AI responsibly and guard against potential attacks.

Without further ado, we present:

When artificial intelligence goes awry: separating science fiction from fact