The resignation of Anthropic’s senior researcher Jacob Coxon, who issued warnings to humanity, also brought the discussions about the out-of-control aspect of technology to the agenda again. In The Question Concerning Technology, published in 1954, philosopher Martin Heidegger did not view modern technology as merely a neutral collection of tools used by man; He pointed out the danger of technological thinking seeing the world and humans as “resources” that are constantly ready for use. Another famous philosopher, Andrew Feenberg, argued that technology is not an independent force that inevitably progresses on its own, but that its design is shaped by social interests and power relations, and sought the solution in the democratic control of technology. These deep-rooted debates are now on the agenda again with a striking resignation from within one of the world’s largest artificial intelligence companies.
“NO COMPANY IS BEING RESPONSIBLE”
27-year-old Jacob Coxon, who previously worked on training new artificial intelligence models at OpenAI and most recently at Anthropic, announced that he resigned from his position on September 8. Coxon, who spent the last three years in two companies, stated that the industry is racing to create super intelligence that can improve itself and accused “No company is acting responsibly.” According to Coxon Many developers of artificial intelligence think the systems could pose an existential threat to humanity by the end of the decade.
A CENTURY-OLD PHILOSOPHY DEBATE
At the center of German philosopher Martin Heidegger’s criticism was the fact that modern technology was changing man’s relationship with the world. Heidegger argued that modern technology reveals nature, and with it humans, as a “ready resource” that can be organized and used. Canadian philosopher of technology Andrew Feenberg carried the debate opened by Heidegger to the social and political field through the Frankfurt School and especially Herbert Marcuse. According to Feenberg, technological systems are not neutral. Technologies can carry social values, power relations, and the preferences of the institutions that design them. On the other hand, Feenberg also argued that with the “democratic rationalization” approach, the direction of technology could be changed by the intervention of users and society.
SAME CONCERN FROM BOSTROM TO HINTON
Oxford philosopher Nick Bostrom has taken the debate directly to superintelligence. In his 2014 book Superintelligence If the goals of a system that exceeds human intelligence cannot be kept compatible with the interests of humanity, a “control problem” will arise. defended. Geoffrey Hinton and Yoshua Bengio, leading researchers in artificial intelligence, have also made clear warnings about the risk in recent years. While Hinton said that it is very difficult to keep systems smarter than humans under control, Bengio stated that he is concerned that the artificial intelligence race between companies and states will push security to the background.
