Work and employment

AI will be able to automate many of the tasks currently performed by humans. It is quite obvious that some professions may disappear altogether, while new ones related to controlling and collaborating with AI will appear. According to one study, the spread of powerful AI could put the equivalent of up to 300 million jobs at risk of automation worldwide, as approximately two-thirds of modern professions include at least some tasks that AI can perform instead of humans. However, this does not mean total unemployment: most professions are expected to be only partially automated, and AI will complement human work rather than completely replace it. Historically, technology has also created new professions, and so too could general AI give rise to entire industries that we cannot predict. People may have to learn new skills, focus more on the creative, managerial, and interpersonal aspects of work, leaving the routine or analytical part to machines. Perhaps in a world of strong AI, the human working day will be reduced, and the productivity of the economy will increase.

Training and education

AI tutors and personalized educational programs can make learning much more individualized and effective. A strong AI will act as an ideal teacher that adapts the methodology to a specific student, explaining any topic in an accessible way. Today, artificial intelligence is used to personalize learning, but imagine that AI with a full understanding of children’s pedagogy and psychology will be available in every classroom. Perhaps curricula will be created on the fly to meet the needs of the student, and AI will be able to immediately analyze the gaps in knowledge and the best ways to fill them. Education may indeed become more engaging, for example, with the help of simulations (created by AI), students will be able to relive historical events or conduct a virtual scientific experiment.

Medicine

In medicine, AI will become an extremely experienced doctor and researcher at the same time. Such a system will be able to instantly analyze patient symptoms by comparing them with millions of cases, make diagnoses with higher than human accuracy, and offer optimal personalized treatment plans. Superintelligence will help to discover new drugs and therapies much faster than now by modeling the interaction of molecules or the effect of drugs on the body. We have already mentioned AlphaFold, which revealed the structure of proteins, and future systems will be able to solve other unsolvable problems in biology and chemistry. Besides, AI assistants can monitor people’s health 24/7 (via wearable sensors) and warn of diseases even before symptoms appear.

Entertainment.

It goes without saying that the entertainment sector will also change significantly. First, AI-generated content will flood the market: movies, music, literature, video games generated or co-created by artificial intelligence will become commonplace. Secondly, strong AI will give rise to new leisure formats. For example, fully interactive movies or TV series where the plot adapts to the viewer’s reactions in real time. Or ultra-realistic video games/virtual worlds (as discussed in the previous part of the article), where you can live another life in a simulation. Personalization of entertainment will reach its peak: AI will know your preferences and create content just for you.

As you can see, no area will remain unchanged. Ideally, a strong AI can bring a “golden age” to humanity, with high productivity, victories over diseases, and intellectual and cultural prosperity. But these same changes also bring challenges, which we will discuss later.

CHALLENGES

Creating a strong artificial intelligence is not just a technical task, but also a major social challenge. Such a technology is likely to radically change the world, so you should think about the risks/problems that may arise in advance. We have already discussed some of them earlier:

Security and control over AI (the problem of controllability). One of the most important questions is how to ensure that super-powerful AI acts in the best interests of humans? If a strong AI learns on its own and re-designs itself, there is a risk that it may develop its own goals that differ from ours. Scenarios where AI gets out of control are popular in science fiction, but they are also considered by quite serious scientists. For example, a superintelligence could theoretically find unforeseen ways to achieve a given goal that would be harmful – here’s a classic example: if you ask it to maximize the production of paper clips without setting limits, the superintelligence could destroy everything around it to get the resources for this paper clip factory. This is an exaggerated example, but it clearly shows why it is important to teach AI to understand our true intentions and have built-in restrictions (primarily, not to harm people). Developing methods to control AI is a priority. Some experts say to first create simplified AI prototypes and test them in safe conditions before letting them into the real world.

Abuse and dangerous use. Even if strong AI itself is controlled, there is still a risk that people will use it for harm. Like any powerful technology, AI can become a weapon in the hands of those with evil intentions – terrorists, criminals, and unscrupulous states. One of the fears is that powerful AI will help create new types of weapons or synthesize a deadly virus.

Confidentiality. A strong AI will have access to huge amounts of data and will be able to do anything with it. This raises the issue of privacy: how to protect people’s personal data? For example, if AI processes all emails, surveillance videos, and medical records, won’t it turn into total control? And another unusual aspect is the rights of AI itself. If we one day create a conscious artificial intelligence, should we grant it certain rights (to exist, to choose actions)? It’s hypothetical at the moment, but philosophers are already discussing whether it would be ethical to exploit a conscious intelligent machine as a slave.

And finally, social inequality/affordability of the technology. One practical problem is who will own and control strong AI, if it is concentrated in the hands of one state or a few corporations? Unequal access to AI could widen the economic gap between countries. Those who have it will have a huge advantage in productivity and influence on literally everything…