Years ago, a stand-up comedy club whose name I can’t remember right now argued that what made the most enjoyable in the world was not “neither the food, nor the money, nor the fact of scratching the back… The pleasure of the world is to be right. That was part of the low-key appeal of robotics : having a cornucopia of pots that would listen to us and agree with us.
However, from the start, science fiction warned us that it wasn’t going to be easy. It should not be forgotten that the first work of fiction including robots, “RUR” by Karel Čapek, already results in dozens of these artificial beings organizing a revolution and destroying humanity. Sorry for the spoiler, but it’s important: especially since there are teams of researchers working on it, on the formation of artificial intelligences to oppose us .
Can an AI beat us in a debate?
The “good” news is that they still don’t do it well enough to convince us in the middle of the debate , but they don’t do it badly. Noam Slonim’s research group has just presented in the journal “Nature” an autonomous agent that can participate in a competitive debate with humans. And this is not a trivial question.
So far, imitation work related to language has yielded mixed results . On the one hand, we have succeeded in making algorithms capable of understanding the emotionality of a sentence; on the other hand, things like summarizing texts or engaging in dialogues turned out to be much more complex.
According to “Nature,” Project Debater (which is the name of the set of algorithms) analyzes an archive of 400 million newspaper articles and Wikipedia pages to develop arguments and counter-arguments for use in a debate with humans. The team put it to the test during a debate on free preschool education and used the transcripts of the same for a jury to jot down the blind statements .
Project Debtor has scored well, although he has yet to win a debate. However, as is often the case, the distances are gradually shortening and these small steps bring us closer to a scenario where artificial intelligences can participate in complex human activities such as destroying the world or, who knows, winning an electoral debate.