June 2018, Feuilleton S. 15
Visionary and entrepreneur Elon Musk and the recently deceased physicist Stephen Hawking warned of the existential threat to human civilization posed by the unbridled development of artificial intelligence. Wolfram Henn, a human geneticist and member of the German Ethics Council, also takes up the concerns of these two prominent figures. Justify this in an essay worth reading in the Feuilleton section of the FAZ on 25.June 2018.June 2018.
His main thesis is that artificial intelligence could become dangerous if it is able to act autonomously, which is what AI researchers are working on! Prof. Henn does not argue as a technophobic skeptic or moralizing warner, but rather as a sober observer of innovation cycles of biological evolution.
Life in its most primitive form has existed for two billion years, early forms of human life have existed for about 7 million years. In evolution – according to the author – a superior species or culture has always ruthlessly converted its developmental advantage into dominance. The human being formed an ego-consciousness " when it came or perhaps had to come to an autonomous self-organization of information processing in the brain from a certain degree of neuronal efficiency onwards."
Computers are still far inferior to humans in our ecosystem, successful only in specific challenges such as chess, Go or Big Data – evaluation. However, the military use of drones and also autonomous driving require autonomous recognition of situations, an (evaluative) situation determination as well as independent action orientation. The development progress in autonomous driving can be read every day in technical journals, and in the medical field and robotics, self-learning systems are showing amazing progress. The topic of conversation at this year's ISC conference and trade fair in Frankfurt is high-performance computers that are entering the exascale age (ten to the power of 18 calculations per second = 1 exaflop). Quantum computers create 20 qubits, their development is just beginning.
Prof. Henn cautions that technical successes should not be viewed in isolation from social benefits and, above all, should not be developed and applied without human control. At this point, a longer quotation is worthwhile: " Only one thing is missing, namely the mechanism of an autonomous further development independent of external control. This is exactly what we are beginning to build into our devices in connection with the much-vaunted artificial intelligence as "self-learning systems" – in the implicit and probably naïve expectation that these autonomous processes will somehow come to a standstill before they can become unpleasant or dangerous for ourselves as the original constructors. Unfortunately, such a gracious self-limitation in the biological. In evolutionary theory, "we humans are the ancien régime of the 21st century". Century".
Prof. Henn did not deny the beneficial developments z in the closing remarks.B. In the medical field, but he rightly warns against educating electronic assistants to autonomy, which empowers them to control their own destiny and then that of humans.
This article reflects the concerns that led members of the association "regional Resilinez" to take an interest in the effects of technology, to get involved to warn against undesirable developments and to develop solutions on a broad democratic and emancipatory basis. All the more gratifying when critical observers like Prof. Henn can have his say. AI – Researchers such as Jürgen Schmidhuber and Ray Kurzweil, while seeing similar developments that Prof. Henn warns, but they don't care if humans still play a role in the long run.