A Twitter bot had to be closed down after it inadvertently sent out a death threat to another bot, prompting questions over responsibility for the behaviour of software.
Technology student "Wxcafe" had created a bot for software developer Jeffry van der Goot, which he said "takes all my tweets I’ve ever made, takes random chunks of it and tries to make new sentences that make sense".
One of the randomly generated tweets sounded enough like a death threat to invite a visit from the Dutch police, though van der Goot claims the tweet was a response to an event written about by another bot, and as not directed towards the other account.
"I had to explain Twitter bots to the police," he said. "And I can’t really blame them for having to take it seriously.
"But apparently I’m responsible for what the bot says, since it’s under my name and based on my words, which I think is an interesting legal angle."
The software developer has since deleted the offending bot.
Christopher Boyd, malware intelligence analyst at security vendor Malwarebytes, said: "Once you load up your bot and send it into the wilds of Twitter stuffed to the brim with words to reassemble, you have little control over what will be said."
The post in question had actually been found by the police themselves, rather than being reported to them, raising questions of how they discovered it.
Commenting on the development, Wxcafe said: "International law isn’t quite up to this I guess."