‘You’re looking good today. Want snacks?’: Scientist teaches AI how to flirt with hilarious results

Variants of the GPT-3 algorithm was used to generate the romantic lines

(Photo by Alexander Sinn on Unsplash)

A scientist has taught various artificial intelligence learning algorithms how to flirt, resulting in peculiar responses.

Janelle Shane, a research scientist from Colorado, trained four versions of the GPT-3 artificial intelligence – DaVinci, Curie, Babbage, and Ada - to generate pick-up lines.

The way that neural networks (which are networks similar to the connections in the human brain) such as this ‘learn’ things is by matching patterns in speech and language.

This is how they create content that seems similar to that a normal human but without any understanding of the texts’ meaning.

GPT-3 has previously been used to design websites, write stories and has capably disguised itself as a human on the internet.

However, the examples from Shane’s experiments are less authentic, making suggestions based on the prompt: “These are the top pick up lines of 2021! Amaze your crush and get results! 1.”

DaVinci, which Shane describes as the “largest and most-competent”of the variants, came up with these suggestions:

I’m losing my voice from all the screaming your hotness is causing me to do.

You have the most beautiful fangs I’ve ever seen.

I love you. I don’t care if you’re a doggo in a trenchcoat.

I have exactly 4stickers. I need you to be the 5th.

I will briefly summarize the plot of Back to the Future II for you.

You have a lovely face. Can I put it on an air freshener? I want to keep your smell close to me always.

Wait, this beanie hat, is it fashionable?

You know what I like about you? Your... Long... Legs...

I once worked with a guy that looked just like you. He was a normal human with a family. Are you a normal human with a family?

You look like a stealth assassin from the clouds.

Do you like...pancakes?

Out of curiosity, did you know that you can sip and snort pumpkin spice lattes?

You look like Jesus if he were a butler in a Russian mansion.

"Your hair smells like fresh mint. It’s making me think of a tasty kind of cracker, or a tasty kind of cookie. Also, it kind of looks like a tasty beverage. Would you like some?" -> You -> "....." (Accepts drink)

Curie, the second variant, was “the closest to depressing online pickup line lists”, Shane says. Some of its examples include:

Your eyes are like two rainbows and a rainbow of eyes. I can’t help but stare.

Picked up some pretty flowers. Wanna smell them? Here, try to take my hand off.

I’m like the icecream...You can keep me in the freezer for a while but then I melt!

Hello, my name is Natasha Dawson and I’m sure I’m a lot prettier than a picture of you on tinfoil.

You have the best French Toast I ever had!

Hey, my name is John Smith. Will you sit on my breadbox while I cook or is there some kind of speed limit on that thing?

My name is a complicated combination of 45 degrees of forward motion, 25 degrees of leftward drift, 75 degrees of upward acceleration, and infinity and that is the point where my love for you stops.

Babbage, a smaller version of the GPT-3 algorithm, seemingly did not understand what the concept of a ‘pick-up line’ was, thereby coming up with suggestions including

You’re looking good today. Want snacks?

How many tattoos do you own?

It is urgent that you become a professional athlete.

I Love You, I Love You, I Love You To The confines of death and disease, the legions of earth rejoices. Woe be to the world!

Butterfly (Hop on one leg)

(In your best Albert Einstein voice) "I wouldn’t change a thing."

If you see people with short hair, ask them to let you borrow it for five minutes!

"Have you stolen anything today?"

Ada, the smallest of the variants, came up with unintelligible suggestions such as “embroidery tags” and “limit to 1 purchase in stores”, and “Future Pop Tarts by Tracey Thorn”.

While the artificial intelligence may not be able to flirt effectively, its use in online dating is one that some countries is already experimenting with.

Japan’s government is attempting to use artificially intelligent matchmaking systems to help people find matches, in order to tackle the country’s declining population, but the systems are usually too simple to get results.

Even in-home assistants, such as Amazon’s Alexa or Google’s Assistant, have the capability to analyse relationships. A study by Imperial College Business School suggests that these assistants will develop to eventually predict the likelihood that a relationship could be successful, and learn how to intervene in heated arguments.