There are people using AI to “invent” their own memories from photographs. And that has opened a philosophical debate

In the photo the kid smiles, abholly happy. His mother, also smiling, hugs him while looking at the camera. The photo could be that of any of us with our mother, but with this happens one thing: that an AI has made it a video. One that immediately has the potential to become a memory.

A memory of lie.

That photo shared it in x Alexis Ohanian, co -founder of Reddit, millionaire and entrepreneur. Ohanian – caught with Serena Williams since 2017 – told in that message how when he created that video innocently “I was not prepared for how I would make me feel this.”

In his family they had no video camera, so he never had a video with his mother. So almost without thinking he used that photo in the newly released Midjourney video generator so that from it generated a video for AI. The result left him astonished. “This is how she hugged me,” he explained. “I have seen it 50 times again.”

But with that message A great moral and philosophical debate was unleashed. One about how something like this can impact us individually and as a society. While Ohanian himself He defended himself Before those who criticized the idea (“I really don’t understand why you wouldn’t use AI for this”) others They explained to him that those memories were not real: “Creating a video for someone loved is not to create a memory of them. You are putting words they never said in their mouth.” As said Another user named Erich Thilow, “is (a memory). But seeing it will make it real in your memory. I am not a fan of something like that.”

Other users took advantage of cinema as the center of the debate. A user named Vanillaelle shared A Harry Potter photogram revealing with a false memory of his parents in the mirror (which has inspired the image of this theme). Another user called Andro toward The analogy with ‘Matrix‘. In that movie, he explained, one to see her wondered why people would want to face reality. “Now he doesn’t have to ask him,” said said user.

The era of false memories

That new capacity of the generative AI Generate false memories It is disturbing, and it could well be part of that concept already studied in psychology. These lies memories did not happen or are the distortion of a real event, and according to experts, such as the American psychologist Elizabeth Loftus, it is possible to induce them such as hypnosis or with techniques such as the essay: repetitions of an event that was confirmed as fantastic. Upon listening or visualizing the same event, the person can begin to remember as if it had really happened.

That idea thrown by this psychologist also collects it Francisco Taberneroclinical psychologist in Puertollano (Ciudad Real). According to him “memories are usually quite distorted”, but for him there is no special risk in that type of process that converts a photo into a souvenir of lie:

“The general experience of memory will be subjective and more or less the same we had. I do not think that a few seconds of movement change the memory that you already had. At the emotional level the memory is all emotion, what causes you remains the same as what you have told to the AI.”

Loftus – criticized by the pseudoscientific concept he created, the “false memory syndrome” – published A study In this regard, together with Ira E. Hyman Jr., and both explained how “memory is always constructive. People create the past based on the information that remains in memory, in general knowledge, and on social demand to recall situations.”

Ethical and moral issues are definitely huge, and to talk about them we wanted to contact more experts. Santiago Sánchez-Migallón (The von Neumann machine) It is defined as a “philosopher of AI” and is usual collaborator of Xataka. For him, “we must understand the emergence of a technology as an opportunity and no, a priori, as something perverse only because of the fact that he treats a delicate issue as, in this case, the death or authenticity of memories.”

Sánchez-Migallón makes it clear that the first thing is to make technology safe and that the user “can differentiate true and false memories”, but assured this, this expert thinks that with this technology it would be possible to help “erase traumas” and even imagine a not very distant future in which it was possible to gain procedural memories or skills: “Could we record in our brain the ability to play the piano or talk?”

Darío Benítezpsychologist and co -founder of PSYCHOFLIX and of the podcast Validlyit is optimistic with AI, but you see “few or no advantage” to this type of application because “if you are reliving an image of a loved one you expose yourself to an emotion that you did not expect and that can rekindle a type of duel that you had already prosecuted.” Something like that, he indicates, can change the perception of your values ​​because this type of lies memories “can make you feel that you lived them and that that effectively happened.”

It is like Ohanian’s own example. For Benítez it could happen that in the video generated the mother looked at the child with a subtle disdain and generated a reaction of the type “yes, it is true that my mother did not treat me at all well”, thus generating a hallucination that connects with another idea that perhaps had of her past, which in that case would be of the type “my childhood was not so good.” All that “would even more entangle things,” says this psychologist.

The danger of being able to mold memory

An technology also has another danger that we have already perceived in many other technological areas. Especially in mobile phones and social networks, which tend to isolate us and catch us in the doomscrolling.

With these videos created by generative, “it could be the case that people I would prefer false lives, renouncing reality“, says Sánchez-Migallón. Despite that, he continues to see it as an opportunity:

“There are already people who spend more time on social networks than in real life, so that in their biography it matters more their world online than their ordinary world. Well, the same would happen here: it is to open the door to another option of life. If there are people who hate their life and prefer to live a false life, why am I going to deny that? And if we discover that living in worlds of invented memories makes us much happier than facing the unpleasant reality? Humanity to suffer because it seems “weird” or “strange” to create false memories? “

For Sánchez-Migallón there is another derivative: memories contribute to forming our personality and “the things that have hurt me have contributed to make me as I am. I do not know to what extent I could erase things I could not create strange incongruities.”

For his part, Benítez explains that “evoking memory is very easy and we know that the mind can make many mistakes.” That is why it emphasizes that baseing our decisions or reaching moral judgments based on those false memories “is dangerous and can create problems where there are no.”

It is something that also seems to coincide Tabernero, who comments on how “there are risk populations” for this type of applications, such as people with post -traumatic stress or with pathological duels. In the latter case, if someone has had a traumatic duel for a loved one that is no longer,

“That tool could rekindle an anchor memory that causes the person to experience symptoms like the first day. At least he will rekind you nostalgia, and you can spend a time of penalty and sadness that maybe you could have saved yourself.”

Benítez does recognize that other technological platforms have helped in the past. Whatsapp and even Tinder help connect people who would not otherwise (or they wouldn’t do so), explain, but this type of technologies – as sadly happens with social networks – can also isolate us. According to him:

“People use virtual environments like those offered by the ‘World of Warcraft’ game because they are safer is already good, but that virtual alternative can make you eliminate the others (real) and also threaten social contact. Something like that can make you live a reality that does not marry and that is not really tangible.”

And just where the reflection of Tabernero also derives, which clarifies that it is very difficult to forecast what will happen, but that here sees two aspects. The first, that of the tweet, “I see it very harmless, almost like a hobby.” The second, much more disturbing, is what can cause “A while of evasion in which you feel good“If you avoid too much, explain, your contact with reality can be much more difficult for not being accustomed to hostile stimuli.

The AI ​​is scheduled not to disappoint you, to always support you, but as Tabernero points out, it can cause the person to “do not develop coping strategies, and those hostile experiences will then be twice as much hostile.”

Image | Warner Bros.

In Xataka | The great mobile paradox in adolescence: they are bad for performance and mental health, but prohibit them in school does not work