With advances in AI, is humanity signing its death warrant? Leaders in artificial intelligence raise the alarm in an online statement about the threats of extinction of the human species with the rise of artificial intelligence. Addressing AI-related risks should be “a global priority alongside other society-wide risks, such as pandemics and nuclear wars,” the signatories write on the AI website. Center for AI Safety, a non-profit organization based in the United States. Among the signatories, there are great figures in the field such as Sam Altmancreator of ChatGPT or Geoffrey Hinton, considered one of the founding fathers of artificial intelligence.
Pascal Picq, paleo-anthropologist, who participated in the work The meaning of tech (Philosophie magazine) where he discusses the major challenges of technology. It helps us see more clearly about the real risks of AI for the future of the human species.
Should we be afraid of seeing humanity die out with the progress of artificial intelligence?
It would first be necessary that the signatories of the forum specify what type of artificial intelligence it is and how it will eliminate humans. The debate is not recent, it started with killer robots and killer drones. It’s very real, they are weapons of destruction, but it’s like the atomic threat. There’s plenty to worry about letting go of technologies that can fire all around depending on their programming. The debate is at least fifteen years old. The arrival of ChatGTP raises the question of general public access to generative AI. Until now, AI was looking for data on scientific or literary knowledge, validated and recognized. The difficulty with these systems capable of creating new content is that they have no scientific validation. But this concern is part of what we already know with social networks. Truth is no longer based on demonstration, it is based on facts or lies being repeated over and over again. The real problem concerns the relationship to knowledge and a shared understanding of the world. This worries me a lot.
The brain is made to work and if it no longer works, it deteriorates
If there is an issue of knowledge and truth, can’t we think that AI will transform humans as we know them?
In my book Artificial intelligence and the chimpanzees of the future (Odile Jacob), I am talking about the “planet of the apes syndrome” which comes from the novel published in 1963. In the dystopian world described by Pierre Boulle, human societies have invented machines which make it possible to produce heaps of things and domesticate the great apes -chimpanzees, gorillas and bonobos-. In the third part of the book, we discover how humanity has been dominated by the great apes. A woman explains: we were so well on the planet Soror, we had machines and great apes who took care of all our needs. Over time, we ceased to be intellectually and physically active. This is the real danger. In our societies, we are already experiencing a real loss of physiological quality of our lives with obesity and physical inactivity, which pose serious health problems on a global level. The human body, like the brain, is made to work, and if it stops working, it deteriorates. We observe an alteration of our cognitive abilities with the use of these machines. It’s not the machines themselves that are to be denounced, it’s the way we use them.
So, somehow, humanity will disappear?
Humanity is not going to disappear, it is going to be profoundly transformed. These machines are capable of solving a lot of questions and we will no longer solicit this brain which needs to function to maintain its capacities. There is also a risk of loss of social ties, of focusing on oneself, of isolation that we already know and which will certainly increase with AIs. The danger concerns above all the disintegration of what has been the foundation of human societies for two million years.
A part of humanity is more likely to find itself in situations of dependence on machines
Experts refer in particular to the loss of jobs. How can the disruption of work contribute to the end of our species?
Throughout human history, widespread technological revolutions have always created more jobs than they have destroyed. It is empirical. Two questions emerge. The first: AI is better than us. If we make machines, it is so that they are better than us. Cars were made to go faster, they didn’t replace us. They destroyed jobs in the horse-drawn carriage world, but other jobs were created in the industry. The second question is: replacement. Machines are not going to replace us, they do tasks better than us. They will not do all the tasks, they will create new jobs. The evolution of humanity is always linked to technologies. Fire, electricity, today AI.
Is a sort of sub-human taking shape?
Part of humanity is more likely to find themselves in situations of dependence on machines, unable to have enough distance to make decisions. This is already partly the case. We are a very plastic species, we can change very quickly from a morphological (size and shape), physiological (body functions, life expectancy, health) and cognitive point of view. Plasticity is great when it is oriented towards objective improvement, it can also go in the wrong direction when the contexts are no longer adequate. And we are already getting to know it.