The future of AI is also the future of the human condition.

Artificial intelligence will radically transform the relationship between work, intelligence and economic value, redefining what it means to be useful, to work and to create value.

The change could go even further: what happens when human capabilities are no longer essential to the economy, creativity or decision-making?

scroll

Too fast for the human pace

AI is evolving faster than our ability to anticipate the consequences.

In just a few years, AI has gone from incoherent text to above-human performance on some difficult tests.

The length of programming tasks that AI systems can complete with an 80% probability of success, measured by the time an expert needs.

010 min20 min30 min2019202020212022202320242025 2019: a few seconds 2019 a few seconds 2025: 30 minutes 30 minutes 2025

This length doubles roughly every seven months.

Source: International AI Safety Report 2026.

Radical change

First, activities are automated. Then, roles. Then, organisations.

01Activities
AI takes over parts of an employee's work.
02Roles
Some positions no longer need to be filled by people.
03Organisations
Companies can redesign their structure around AI systems, without people.

Beyond work, the same dynamic can affect creativity, expertise and human participation in decisions. When a system can produce, decide or coordinate more efficiently than a person, human capability risks becoming irrelevant not only economically, but socially as well.

The New Social Contract

An economy, in a society that no longer needs people.

AI can produce more and more, with less and less effort, at lower cost.

Less dependence on people. Less power for people.

The Intelligence Curse describes a fundamental risk: if the economy no longer depends on people's contribution, those who control AI, or governments, could have little interest in investing in their education, health or social protection.

Technological progress does not guarantee human progress.

Source: Luke Drago & Rudolf Laine, The Intelligence Curse (2025).

The three risks

What do we risk losing in a world dominated by AI?

01
Control
AI systems can act in unpredictable ways, pursue goals incompatible with human intentions, or become difficult to control.
02
Democracy
AI can amplify manipulation, surveillance and the concentration of decision-making power, weakening people's ability to take part in public life and to influence the decisions that affect their future.
03
Equity
AI's capabilities and benefits can become concentrated in the hands of those who own the technology and infrastructure. At the same time, human work, creativity and expertise can lose their economic and social value.

These risks are distinct, but they can reinforce one another. Together, they call into question not only how we use AI, but also the place people still hold in society.

The choices

Building more capable AI is not enough. We have to decide what role we give it.

AI safety is about control over systems, protecting democracy and the fair distribution of capabilities and benefits. It also means keeping a real role for people in work, creativity and decision-making.

AI safety is not only the responsibility of engineers or experts. It is a problem for everyone who will live in a world transformed by this technology.

e.

educativ.digital

Understand AI. Understand what it changes. Take part in the decisions that matter.

Free introductory courses, articles and analysis, or a newsletter for what matters. And consulting for organisations.

Explore the resources

The future of AI depends on the choices we make today.

Understanding change is the first step towards influencing it.