Could a robot make you laugh?

Artificial intelligence researchers are trying to make machines tell jokes. It's not going very well.

Sign Up

Get the New Statesman's Morning Call email.

Our brains are very complex, sure, but we’re not the transcendent beings we think we are, despite what that pinkish-beige thing encased in your skull would have you believe.

It’s hard to fathom how sentience is the result of a meshwork of nerve cells and chemicals. But the more we learn about our brains, compare them with other non-human brains and replicate them in artificial form, the less special we appear to be.

Artificial intelligence (AI) researchers are testing the boundaries by seeing if the cognitive skill to provoke laugher and provide amusement can be placed into machines. Basically, they want to see if robots can be funny. Crazy? Sure. This is mostly because humour is dependent on multiple parameters, many of which are internal and subject to change – what might be funny today may not be funny tomorrow.

Linguistics and psychologists believe good jokes all share the same properties – they amuse us – so systematic analysis ought to reveal them, right? Well, erm, not quite.  

Computer scientist Dragomir Radev of the University of Michigan and friends at Yahoo Labs, Columbia University and The New Yorker have been studying cartoon captions to see if humour can be arithmetically expressed in computers. Radev and co’s study is published in arXiv.

The New Yorker’s famous cartoon caption contest has been running for more than a decade. Each week, editors publish a captionless cartoon and more than 5,000 readers submit a funny caption. The editors pick the top three and ask readers to choose the funniest.  

In the paper, the authors of the study take a computational approach to determine what differentiates the funniest captions from the rest. They use a number of standard linguistic techniques to rank all 300,000 captions. Criteria include the level of sentiment, whether the captions were referring to people, how clearly they refer to particular objects in the cartoon, and so on.

Radev and co then took the highest ranked captions and compared them to the gold standard: the captions New Yorker readers chose as the funniest. This was done by crowdfunding opinion using Amazon's Mechanical Turk, a place where companies perform tasks that computers are currently unable to do. 

Based on this approach, it’s easy to imagine a computer capable of churning out the best caption. But the researchers are a long way off from achieving this. A more ambitious goal would be to have the machine write the best caption for a cartoon – good luck achieving that.

I ask Radev how we might build funny robots, if ever. He replies: "Easy – leave a few of the screws loose". 

Perhaps it should come as a relief that making jokes is one more thing a human can do that a computer can’t. So maybe we are a little special after all, at least for now. 

Radev and co are making their database of captions accessible to other researchers. If you would like to build the first funny robot then you're welcome to it.

Tosin Thompson writes about science and was the New Statesman's 2015 Wellcome Trust Scholar.