Artificial intelligence (AI) refers to technologies that mimic human abilities such as strategic thinking and language, opening up new possibilities for our future. On this page, we answer the most important questions.
Artificial intelligence (AI) is the endeavour to transfer key aspects of human intelligence to machines. As there is no binding definition of the term "intelligence", a universally valid definition of "AI" has yet to be found. But what can be said in any case: Similar to us humans, computers should use data and experience to learn, make judgements and solve problems independently - in order to perform tasks better and better.
Everything you need to know about artificial intelligence, how it works, its uses and how to check its safety.
Machine learning is now on everyone's lips. The concept behind it: Instead of being meticulously programmed for the respective application, the machine learns to recognise patterns and regularities in images, for example, with more or less human guidance. The different ways in which AI is trained, what can go wrong and how the so-called bias can be prevented, we have summarised.
The more responsibility we transfer to AI, the greater the demands on the safety of the machines: If we want to let artificial intelligence control our cars in the future, it must function reliably at all times and must not allow itself to be manipulated. Read here about the criteria AI can be tested for the autonomous car.
Computers can do everything better and will eventually make us humans superfluous - this worrying suspicion seems to be becoming increasingly true: Artificial intelligence has already shown chess grandmasters and Go professionals their humanoid limits. In any case, one thing is certain: AI can perform various routine tasks faster and more persistently than human labour. Increasingly, however, AI is also being used in artistic fields. In this article we show how AI is used in music and painting, for example, and why its creativity has narrow limits.
But even with its eponymous "intelligence", things are not as advanced as often assumed, says Henning Beck. In the interview the neuroscientist explains the extent to which small children make every supercomputer look old and why computer science is still struggling with the complexity of the human brain.
Physicist and neurobiologist Christoph von der Malsburg also believes that AI is still a long way from human intelligence. The renowned AI researcher provides detailed information on the conditions under which it could develop "real" intelligence, the risks involved and why we already need to regulate AI today. provides detailed information.
AI can play a major role in more and more areas in the future. And we need to agree today on where we draw the safety boundaries, which systems we want to authorise and which not.
Vasilios Danos, expert for AI security
TÜVIT
Artificial intelligence recognises our face on our smartphone or traffic signs on the side of the road, navigates us past traffic jams through the city, gives us tips on selling shares or the next series on Netflix and understands our voices with Siri, Alexa and the like. But its potential areas of application are far from exhausted. AI can also improve the climate, support doctors with diagnoses and make transport and make transport more efficient.
Without question, the climate crisis is real. We all continue to produce far too many greenhouse gases. The consistent use of AI could reduce global CO2 emissions by around four per cent by 2030, according to an estimate by Microsoft and the consulting agency PwC. However, artificial intelligence also offers great potential for the Federal Ministry for the Environment to protect the climate and the environment: since 2019, the ministry has been funding projects that aim to overcome ecological challenges using AI. In fact, there are already a whole series of approaches to use artificial intelligence to improve water quality, stop the loss of species or grow crops more sustainably.
Our cities are groaning under traffic jams, exhaust fumes and traffic noise. How AI can be used to predict pollution levels traffic lights can be controlled more intelligently, delays can be reduced - and what is behind the intelligent bollards being trialled in Karlsruhe.
At the Associated Press news agency, machines have also been writing texts for several years: robo-reporters write reports on quarterly balance sheets or share price developments at companies. However, algorithms can also create football reports, weather reports, stock market news or product descriptions at the touch of a button.
