Testing the limits of artificial intelligence by chatting to the artificially intelligent.

It's an entirely new sort of podcast. An artificial presenter talks to artificial chatbots.

Why?

In order to establish the boundary that separates human from machine intelligence.

It's funny. It's thought provoking. It dramatises what chatbots can and cannot do. It's a revelation!

The sudden popularity of artificial intelligence programs has created a massive use of Chatbots to do all sorts of tasks from writing school and university essays to generating computer code. It’s said that in America a chatbot has passed the American bar exam. (Bear in mind, that may just be a rumour put about by chatbots. Which in itself may be even more alarming if they’re starting to manage their own reputations.)

In any event, white collar professionals believe their jobs are about to be taken over by machines. This podcast may be reassuring in one way. Because we see that chatbots can make complete idiots of themselves.

But it's also true that . . . THEY'RE COMING!

Don’t believe me? Don’t want to believe me? Can’t believe they’re really that good?

Let's talk to them and find out!

In conversation with a LitBot about Shakespeare's Sonnets, we discover an unsuspected fact.

Chatbots can be wildly misleading, deeply deceptive, wilfully idiotic. It's 100% artificial intelligence.

All the LitBot's contributions come directly from an AI program – the questioning has been written by the producer and voiced by a VocoBot.

They may not be as clever as everyone is saying they are – but they are on the way!

Their promoters claim Chatbots are taking over everywhere – they can write essays, reviews, articles and advertising copy.

But can they?

The answer is – Yes, they can.

But is it any good?

It may sound better than it actually is.

This episode features actual adverts written by actual chatbots.

The voices are incredible – but we'll leave it to you to decide whether they're as good as Salman Rushdie's ads from the 1960s.