Let’s call her Anna. Her parents have an average income. When Anna goes to primary school, she already has a digital assistant, controlled by AI. She chose her assistant’s name herself, with the help of the assistant. Her name is Mia. Mia obviously needs to get paid. This is done via a subscription. The monthly sum is small, but it is for the rest of Anna’s life. The benefits are unlikely. After all, Mia screens Anna in a playful way. She knows what Anna likes to do and what her talents are.
That information becomes crucial for Anna’s future. AI has taken root in this society that prefers efficiency, cost savings and time management. Even before Anna goes to secondary school, her digital assistant can train her on what she is very good at and what she will improve on. Anna appears to score reasonably well in languages, but she has formidable organizational talent, a lot of empathy and is solution-oriented. The assistant constantly helps her in this. Anna rarely feels like she is going to ‘school’. For her, school is what she does in a group. There is no longer a real school building. Teachers guide a process of awareness about living together, collective mental hygiene, social interaction, moral concepts, critical group formation.
The collective is therefore externally guided, the individual hardly.
After all, there is no teacher who knows Anna so well and helps her as well as her digital assistant.
Moving avatar
Everything that Anna communicates to Mia forms an inexhaustible database, also endlessly readable by others. Mia helps manage Anna’s consumption patterns and all her preferences. There is a reasonable chance that Mia will help find a partner for Anna. She will save her from a lot of pitfalls in her life, but not from everything. After all, people often cause chaos in their lives and there is little Mia can do about that.
Nevertheless, she finds a partner, has two children and eventually a balance is found between family and work. When Anna feels a loss, Mia teaches her to meditate. She was also the first to save Anna from a possible physical ailment. No doctor can screen another human being so specifically. Anna knows that if she listens to Mia in terms of food and exercise, she will be able to enjoy a long life, within the set parameters determined by her hereditary burden, stress management and the environment in which she lives, such as air quality, presence of greenery, and so on.
While Anna is dying after that long life, she is surrounded by her children and grandchildren. They are sad because her body is giving up. But they also know that Anna will never disappear completely. After all, Mia becomes Anna’s ethereal body. Anna can always be summoned again after her death. Mia will be able to give advice as Anna, simply listening to her descendants, answering with her voice and with a moving avatar of hers. For example, the children will be able to call up Anna as a young mother and ask her questions about it. Everyone knows that. Dead will never be truly dead again. As long as the subscription remains paid.
Yearning for confirmation, I ask ChatGPT to think about this future Anna. I am told that it is important “to recognize that the future of AI is not completely predictable and that there are diverse perspectives and possible outcomes. It is wise to think about the possible implications of AI and continue to actively debate how we can use this technology in a responsible and ethical way.”
Okay, but what does Anna look like, for example? “Anna could be physically described as a woman of average height, with a slim but healthy physique. Her face is lively and expressive, with warm eyes that radiate curiosity and empathy. She has a bright smile that reflects her friendliness and openness. Her dark hair falls in soft waves around her face and reaches halfway down her back. Anna has an energetic appearance and her attitude exudes confidence and determination.”
Which actress can she be compared to? “Emily Blunt.”
Unrestrained superintelligence
What I have written is, in my opinion, a very limited, far too simple approach to what the impact of AI can be. On Facebook I am told that my scenario has already been depicted or explained so many times in novels, film scenarios and TED talks. Anyone who tries to follow AI is a layman who feels confined within the parameters of a sci-fi world that could soon become reality. I ask my buddy Jonathan Berte, founder of the AI platform and company Robovision, to take a look at my text. Of course, within a few years we will have an AI assistant of our choice, with the voice, avatar and style of our own idol, he says.
But who will control these assistants: big companies or the open source community? This question is extremely topical, and the technology giants are diametrically opposed to each other on this issue. Yann LeCun of Meta, a leading AI expert, passionately advocates open source. According to him, the risks in terms of security and individual privacy do not outweigh the danger of superintelligence being centralized among a few, such as the teams around Sam Altman of OpenAI and Satya Nadella of Microsoft. Imagine that advertising is no longer recognizable as such, or that political parties pay companies to filter information or send choices via your AI assistant. At the other end of this spectrum, Sam Altman warns of the dangers of an unchecked superintelligence. What if this makes it easier for terrorists to obtain explosives? He therefore proposes to keep it closed and have this checked by companies.
After this explanation, Jonathan advocates a vision of openness, towards a new Linus Torvalds (founder of Linux) who leads us to general artificial intelligence (AGI) with a transparent system that allows everyone to contribute. In the 1990s we faced a similar choice with operating systems such as Unix and Windows, which suddenly had a lot of influence on our lives. A group of unpaid pioneers, the modern Beggars, then laid the foundation for what is now the most widely used operating system in the world.
Future stories
This future oracle of Delphi, which will contain all human knowledge and secrets – he writes to me – may also help to make social decisions more transparent and efficient, such as an environmental impact report or a decision on a new bus line. And if we are really ambitious, why not replace a country’s measure of success – currently expressed in gross domestic product – with an indicator that better measures the well-being of people and nature? These are the issues we face as a society, and the choices we make will affect us for a long time…
Jonathan teaches me a thing or two here, not only about the challenges we face, but also about my way of thinking and feeling that I share with many of you. When new technology is introduced we go through an intense emotional process that usually involves doom or euphoria, fear or optimism. Anna and her assistant Mia appeared in my imagination. But that image has been subconsciously influenced by other stories in different formats.
Stories are contagious. Our stories about the future are usually based on an intensification of current parameters. We like some drama and alienation, a shock or a vision, and at the same time we are loose with ChatGPT. We simultaneously conjure up a totalitarian state and a digital lazy country, without necessarily taking note of the real test of strength behind this technology. We think like prophets of doom, future subscribers or victims. We think too little as citizens who live in a present where the future is not yet certain. It requires an active debate, ChatGPT also told me.
