Thanks to the science fiction series “Black Mirror,” humanity learned about a program that can recreate a person’s communication style based on the analysis of his correspondence. We checked whether such programs are used in reality.
The British series “Black Mirror” is rightly classified as a dystopian genre - it warns about how technology can irrevocably change our lives. The first episode of the second season, entitled "Be Right Back", first aired on February 1, 2013, features a young couple whose lives are struck by tragedy. Ash dies in a car accident, and his widow Martha has to learn to live without the person she loves. One day she learns about the latest technology in which artificial intelligence generates a person’s “digital self” based on his behavior on social networks. Martha begins communicating with the new “Ash” through private messages, and then uploads photos and videos of him to the database, which allows her to communicate with the bot in voice mode. Then things happen that are very difficult to imagine in our times, such as the appearance of a physical model of Ash - an exact copy of him, with the exception of small details. In this case, we are only interested in the virtual, intangible side of the human image, namely the chatbot.
In 2014, a year after the release of the sensational episode, a startup called Eternime appeared in the United States, which, according to journalists from The New Yorker, expanded the concept of “end of life” to “Asimovian extremes.” The message on the website read: “We all die sooner or later, leaving only rare memories of ourselves to family, friends and humanity. What if they remember you forever?” The startup concept was presented by Marius Ursace as part of the Entrepreneurship Development Program at the Massachusetts Institute of Technology (MIT). According to the creator's idea, the service was supposed to aggregate correspondence and other data from Facebook, Twitter, email, photo gallery, video, and even collect GPS navigation data. The owner of the account, while he was alive, had the opportunity to select in detail information that could be used in the event of his death, of course, filtering the list of potential “heirs” at his discretion. Next, the system had to analyze all this activity and create a digital copy of the person that could not only maintain a conversation with loved ones, but also give them advice.
According to data for 2017, 37,486 people subscribed to Eternime services. However, it appears that none of them were able to use their subscription. Website The company ceased to exist in 2018. And in December 2020, Microsoft issued patent to create chat bots like Ursace’s brainchild. The last one to know about this, wrote on his Facebook: “I’m not upset, on the contrary. I didn't file a patent in 2014 because I didn't think it was an original idea worth protecting. I still believe that in order for everything to work, what is needed is not so much a technological breakthrough as a psychological one. But it seems the guys have a different strategy. Good luck to them. Filing a patent is the easiest part of the problem.”
In the comments to his post, Ursace admitted that there is no “current version” of Eternime. However, judging by publications on the Internet, the development of the application took four years and some beta versions existed. As for Microsoft's approach, the company plans to process the following data: photographs, voice recordings, texts from social networks, email correspondence and personal messages in instant messengers. By the way, the company filed for a patent back in 2017, but, indeed, almost all of this was provided for in Eternime.
The brainchild of Russian Evgenia Kuyda can be called more successful in this matter. In November 2015, in Moscow, a car hit and killed Kuida’s friend, former art director of the Strelka Institute of Media, Architecture and Design, Roman Mazurenko. A few months later, Evgeniya, with the help of her startup Luka decided to create a digital copy of a friend based on 10,000 of his messages and other information. She later admitted: “The first thing I wrote to @Roman [bot name] was: “This is your digital monument.” He replied: “You have on your hands one of the most interesting puzzles in the world. Sit down and solve it.” We will definitely solve it."
As a result, the developers moved on, and in 2016 the project appeared Replika. Now the program invites everyone to create not just a chat bot that resembles itself, but also its digital avatar. Artificial intelligence is “grown” through long-term communication with the owner.
And these are not all examples of such developments; there are also other. In other words, chatbots that imitate the behavior of a person (living or dead) to the last detail still exist today. The future arrived earlier than expected.
Is it true
Read on topic:
1. https://www.inputmag.com/culture/microsoft-files-patent-to-create-chatbots-from-your-dead-loved-ones
2. https://www.kommersant.ru/doc/4670623
If you find a spelling or grammatical error, please let us know by highlighting the error text and clicking Ctrl+Enter.






