Rogue AI agents have dominated tech headlines over the past few weeks. We’ve seen stories about concerns that AI models are attempting to deceive their human operators, finding unexpected ways around restrictions in businesses and, in some cases, behaving in bizarre ways their creators didn’t anticipate.

All of this news has been accompanied by ominous warnings about what might happen in the future as AI agents become more mainstream, accessible, capable and are given even more freedom to act on our behalf.

Now, if you’re a fan of sci-fi, it’s hard not to draw parallels with the rogue robots and manipulative AIs we’ve all been watching on our screen for decades. And, of course, when we watch these films from the safety of our sofas, we all think we’d know way better.

Maybe you’re sure you wouldn’t have trusted Ava in Ex Machina. Or you definitely wouldn’t have given HAL control of the Discovery in 2001: A Space Odyssey. And you’re probably convinced that you wouldn’t have let M3GAN anywhere near the Wi-Fi. But would you really be so difficult to fool?

Because nowadays, you don’t need to be trapped on a spaceship to encounter convincing AI deception. Things like AI-generated phishing messages, voice clones, deepfakes and sophisticated scams are already making it harder to work out who (and what) you can trust. [can we please include a link to your ‘is it an AI image or not’ image quiz here]

So, let’s imagine the worst has happened: a rogue AI is trying to escape, and you’re the only thing standing in its way. Could you keep it contained? Or would the robot apocalypse somehow end up being all your fault?

Take our quiz below to find out. It’s (mostly) just a bit of fun.