Hannah Fry explains why you should say please and thank you while talking to AI
Good manners don't cost a thing - and according to Professor Hannah Fry, a little kindness goes a long way with AI.
The mathematician, 42, shared some instructions on how she believes you should talk to chatbots to 'get the most out of' them.
She reckons that there is 'actually a really good reason why' you should be polite, before joking that it's 'not just so that AI will spare you when the apocalypse comes'.
Fry, who starred on the latest series of Celebrity Traitors but was banished this week, is somewhat on the fence about how ingrained artificial intelligence has become in everyday life.
Advert
It would be fair to say she's cautiously optimistic about the revolutionary technology - as although she believes it is 'incredible and world-changing', she's aware that it's not all roses either.
In her latest column for New Scientist, Fry said: "I know that there is some inevitability to machines outpacing the humans, and I know that at its best, AI has a great deal to offer. But at its worst, AI is an engine of extraction."
But as AI is apparently here to stay, the Cambridge professor reckons you're best off not treating it with contempt.
In fact, in a video shared to her YouTube channel in March last year, she strongly recommended saying 'please and thank you' when conversing with a chatbot.
She explained that in her opinion, there is 'only one important thing to remember' when you're giving AI a prompt - that the systems have become 'really, really, really good at role playing'.
"So you shouldn't talk to it like a Google search, or like you're trying to extract information from it as though it's Wikipedia," Fry said.
What Fry recommends
She went on: "Instead, you should talk to it as though it's an improvisational actor that could be any character imaginable.
"The reason for this is that large language models, they don't store facts like it's a database. They generate responses dynamically based on having read everything that humans have ever written and the prompt that you give it.
"That means that AI, the stuff we have now, it doesn't have a stable identity. It doesn't have a fixed worldview, it doesn't have personal beliefs. So if you prompt it in a particular way, it will respond in kind."
Fry pointed out that what you give is what you get when it comes to the likes of ChatGPT, as it mirrors your persona.
"If you try and prompt it as though it's a Shakespearean bard, then it will give you a flowery response," she went on. "If you try and prompt it as though it is an extremely effective and smart scientist, it will give you the relevant response. But if you prompt it as though it is an encyclopaedia, it's going to try and sound like one. But it's still just performing a role and you have effectively just restricted what it can do."
The maths whizz instead suggested that people should 'imagine that they are a film director and that you have a character in mind', before prompting your AI chatbot 'accordingly'.
Offering an example, Fry went on: "So don't say, 'Give me three interesting facts about science'. Say you are a world renowned scientist with PhDs in biology, chemistry and physics and your nephew says that science is boring.
"You only have a few minutes and you've got to give him counterexamples. What extraordinary stories do you use?
"And here is the big idea - if you're working with a character and you want that character to eagerly help you - whoever they are - it's probably better to be polite."
According to some folks in the comment section of the clip, her advice was 'spot on' - as when they adapted their queries based on Fry's instructions, one person said the answers they received were 'hugely different'.
But others weren't so keen on the idea of treating AI with the same respect as a fellow human being.
One person said: "This is a PSA: it can't be bargained with, it can't be reasoned with, it doesn't feel pity, remorse or fear."
Another wrote: "I think being respectful to any system is a good idea from the standpoint of practicing good manners in all things you do anyway. Not to mention presenting a positive model for these data complexities to build off of."
A third added: "I refuse - in fact, I see better responses when I swear at it."
A fourth commented: "Thank you for that narrative, I never thought to use that approach."
And a fifth added: "I'd also add being polite is absolutely free and makes the world a better place."
Fry released a BBC documentary earlier this year which explored the epic rise of AI and the impact it is having on people, titled AI Confidential with Hannah Fry.
Topics:Â Artificial Intelligence, Technology
Olivia Burke