Every good love story has a moment in which the precious ingénue, blind to the complexities of the world, misinterprets the lover’s move. Sally mistakes Harry’s interest for friendship. Romeo, believing Juliet to be dead, poisons himself. The folly of love is not so much about what we do when we are flooded with feelings, but what can happen when we have incomplete data. This is perhaps why a crop of new apps have arrived, harnessing the powers ofartificial intelligence, to offer relationship advice.
One of them,Mei, is billed as a “relationship assistant.” The Android version of the app, which arrived last September, parses text conversations to estimate the compatibility and personality of the individual you’re chatting with, scoring along five traits: openness, emotional control, extraversion, agreeableness, and conscientiousness. The iOS version, which debuted this weekend, has a singular function: to suggest the probability, on a 100-point scale, that the contact is romantically interested.
It costs $9 to buy 100 Mei credits, the amount required to analyze a single conversation. (Larger credit packs come at a discount; you can get 500 for $40 অথবা 1,000 for $70.) Right now, the app can only analyze conversations from WhatsApp, which conveniently lets a user export a chat log. Once a conversation is whizzed over to Mei’s servers, it’s crunched through a series of algorithms that search for clues.
I ran several of my WhatsApp chat logs through the analyzer. Mei needs at least 1,000 words to perform its diagnostics, which disqualified several conversations, including the one with my actual boyfriend, who was begged to text me exclusively on WhatsApp for a few days. Others cut the mustard. One conversation, with an Israeli soldier I’d met onBirthright, returned a 24 percent likelihood of romantic interest. That seemed about right. Another conversation, with someone I had briefly dated, scored slightly higher—but even then, only a 43 percent likelihood, despite some R-rated chatter. The only person Mei suggested was likely to have romantic feelings for me was my oldest childhood friend, a gay man.
Not following the logic, I reached out to Mei’s creator, Es Lee. Lee began tinkering with a program to measure romantic interest after watching a clueless friend get ghosted after a date. Lee took his friend’s phone, scrolled through the texts, and saw that his friend had misinterpreted the conversational subtext. “It’s almost like texting body language,” he says. “Do you wait to reply, or do you reply immediately? Do you use exclamation points? Do you double text? I thought a lot of that could be done with algorithms. It felt like a natural thing to do.”
Lee’s first app, called Crushh, promised exactly that. The “texting relationship analyzer” offered a romantic interest score on a scale of zero to five, as well as insights on the power dynamics in a conversation (i.e., who likes who more). It also prompted users to say a little about each repartee: How old were the people in the conversation, what were their genders? Was the contact a colleague? A spouse? A crush?
The latest onartificial intelligence, from machine learning to computer vision and more
Lee says the app processed “hundreds of thousands” of these conversations, many of them self-labeled with those context clues. That provided a hefty data set of what real text conversations looked like, across various demographics and in different types of relationships. Some of the patterns were obvious—a person who says “I miss you” early in a conversation likely has the feels—but others were more Delphian. “Based on the data, people who have romantic intent use the words ‘night’ and ‘dream’ a lot more,” says Lee.
Other apps have used similar models to juice up sales pitches, advise employees on messaging the boss, or generate context-specific email replies. Boomerang, a plug-in for Gmail and Outlook, makesan AI toolthat proofreads emails and suggests ways to improve them before you hit “Send.” An app calledKeigocombines “advanced psychology” and “cutting-edge AI” to determine the personality of a person based on their emails or tweets, and then provides helpful suggestions on how to approach them.
Like any good assistant, Keigo can slide deftly into many situations: to prepare for the job interview, to win the second date, to better understand a partner after a big fight. But Teemu Huttunen, Keigo’s managing director, says people are mostly using it for love. “To be honest, we were hoping that people would use this in other forms than just dating, but the dating one is the most obvious,” he says. “When you have a Tinder match and you agree to go on a date, the next step is that you would have to say something interesting.”
The app borrows a model from IBM’s Watson, which performed aseries of studiesto map basic personality characteristics onto peoples’ output on social media. IBM’s version translates tweets into its own “big five” traits: openness, conscientiousness, extraversion, agreeableness, and neuroticism. Keigo uses a different framework, based on Meyers-Briggs’ personality assessments. Feed it a snippet of text and it’ll deliver recommendations on how to talk to someone.
By way of demonstration, Huttunen showed me a graph that had mapped my tweets against Oprah Winfrey’s. The insights suggested that Oprah and I are 77 percent “compatible,” and that in a conversation with her, I’d want to emphasize teamwork, my “future journey,” and intuitive reciprocity. (Later, Huttunen would send me an email that referenced our “inspiring” phone call, and I would wonder if Keigo had planted that choice of words.)
All of these apps require a real suspension of privacy—they are, after all, parsing intimate conversations. Lee says Mei anonymizes all of its conversational data, and allows users to scrub their uploads from the company’s servers. By way of caution, the app also displays this pop-up before you upload anything: “In order for Mei to give you analysis on your conversation, the conversation history needs to be uploaded to our servers. If you are not comfortable with this, PLEASE GO NO FURTHER.”
For the intrusion, Lee seems to think the payoff is enough. Right now, Mei is a novelty crush analyzer. But he likes to think about what might happen in the future, with a much bigger data set. “I could go, ‘OK, this is a crush, but what type? Are you just flirting? Are you married? You might be able to start building models for those things,” he says. “When you have enough data, it’s almost like an encyclopedia of people.”
In this dream world, text-analyzing apps can do more than just tell you who to ask out. They are bellwethers of human communiqué, a key to unlock the hidden mysteries of the people around us, a way to become better lovers and friends.
But in reality, they seem to signal only our clumsiness in talking to each other. Whether a text analyzer reveals anything real or not, using one seems to offer a false sense of predictability and a semblance of control over otherwise messy human relationships. Does the emoji mean it’s true love? Did the double text ruin the mood? Am I doing this right? The answers, displeasingly, never live in an app. The guidance there is about as useful as a deck of tarot cards.
More Great WIRED Stories
- If computers are so smart, how comethey can’t read?
- xkcd’s Randall Munroe on how tomail a package (from space)
- Why “zero day” Android hacking nowcosts more than iOS attacks
- This DIY implant lets youstream movies from inside your leg
- I replaced my oven with a waffle maker,and you should too
- 👁How do machines learn? যোগ, read thelatest news on artificial intelligence
- 🏃🏽♀️ Want the best tools to get healthy? Check out our Gear team’s picks for thebest fitness trackers,running gear(includingshoesএবংsocks), এবংbest headphones.