Stay in view of the free updates
Simply subscribe to artificial intelligence Myft Digest – it is delivered directly to your inbox.
Two topics dominated the annual meeting of the World Economic Forum in Davos last week: Donald Trump and artificial intelligence. Of the two, the latter was more interesting and certainly more important. A lot of attention is allocated in the discussion of Deepseek, the Chinese surprise. However, we just learned that knowledge differences: no country will monopolize these new technologies. He surprised these markets. With new technologies, such surprises are not surprising. But it does not change the big question, which is what the advanced intelligence of the machine means to all of us.
Humans are both social and smart. This combination is the “killer application”. They were allowed to control the planet. Human intelligence invented the general purposes that formed the world, from taming fire to the creation of computers. However, with computers that think, this may change. “The man is nothing but cane, and he is the most weak things in nature, but he is a reaging.” Is this uniqueness approaching now?
In Davos, I attended two adequate discussions of bonuses and risk of progress in artificial intelligence. One of them was an interview with Sir Dimis Hasabis, co -founder of Google DeepMind and a Nobel Prize for Chemistry Prize, written by Rola Khalaf, FT Editor. The other was an interview with Dario Amani, founder and CEO of the Ashtarubur and his book author Love machinesWritten by Zani Menton Hadiz, Economist Editor.
The interview with Hassabis confirmed the recent developments in our ability to conduct scientific analysis, especially in biology. He said that more than 2 million researchers use Alphafold, as DeepMind developed. “We have stabbed all the known proteins for science, every 200 million. [T]It is the thumb rule. Does a doctoral student take a fully to find one protein structure? So 200 million took one billion years of doctorate. We have given everything to the world, for free. decade.
Widely, Amodei, we can imagine artificial intelligence as a “country of geniuses in the data stage”, one who may have been the Chinese have made them cheaper than before. However, are these geniuses really? My test will be whether Amnesty International can produce Einstein’s general theory of relativity.
It seems reasonable to have the effect of problem solving, whether it is the level of “genius” or not, noticeable. Among other things can accelerate improvements in knowledge, productivity growth and prosperity. Both are desirable. In recent decades, the increases in the “productivity of total factors” – the best scale of technical progress – have been modest. Moreover, huge numbers still live in extreme poverty, in a frustrating way, slowing progress.

However, it is also clear that accelerated progress can create difficulties as well. The labor market structure may change widely, for example, in this case, a sharp decrease in demand for workers whose assets are trained, but largely routine. The expectations of these effects differ. The 2023 paper written by Eric Bringgloffson and Gabriel Unger notes that, as it was true throughout the computer revolution, the effects on productivity may be modest. However, this time it may be different, with increased productivity, but it is significant economic and social changes in return. Again, depending on how society responds, a successful artificial intelligence may lead to “technology”, with greater concentrations of wealth. The invention of huge numbers of new treatments may significantly increase healthcare costs and also costs to deal with lives that extend to a large extent even if they are a healthier balance. Are people ready to live alongside their great ancestors? Thus, it seems that good things may create real challenges.

Beyond that, the development of the imagined artificial intelligence creates great risks. How can one control its use by rogue actors, including hostile countries, terrorists and mass murderers? What are the moral provisions that one allows Amnesty International in the war? How can one control the use of artificial intelligence in monitoring? Does the “older brother” watch us forever? Again, what do we do about fake manufacturing and fake news? How does freedom survive all these threats?
Hassabis is clear that we need effective global borders to use artificial intelligence. In the era of broken international cooperation and contempt for the idea of ”bases -based international system”, will China and the United States together make artificial intelligence safe? It seems unlikely, not least because they have different views on how to use these techniques.
In 2015, a generally skeptical article on (modest) influence was written on the productivity of new technologies. In the next few years, it may eventually prove a mistake. However, I also noticed that if we were dealing instead of “exclusivity” – artificial intelligence that goes beyond all human intelligence – everything should change.
One of the great ideas in Frank Herbert Sand dunes The series is that in the distant past (our future), humanity has launched a successful jihad against the machines that are thinking. After that, humans had to become super. A pioneering figure explains that “humans have put these machines to calm our sense of beauty, and the necessary self, which we have issued living rulings. Of course, the machines were destroyed.”
This anxiety may prove wise. But I am realistic: artificial intelligence outside the Bandura box.
Martin.wolf@ft.com
Follow Martin Wolf with Myft On twitter
https://www.ft.com/__origami/service/image/v2/images/raw/https%3A%2F%2Fd1e00ek4ebabms.cloudfront.net%2Fproduction%2F487b016f-d1d3-449f-950a-0bd93fa189d7.jpg?source=next-article&fit=scale-down&quality=highest&width=700&dpr=1
2025-01-28 17:43:00
#learn #live #machines