Pope Francis sees killer robots, autonomous drones and similar weapons systems as a major threat. “No machine should ever choose to take a human life,” he stressed in his eagerly-awaited keynote address on artificial intelligence (AI) on Friday evening at the G7 summit in Borgo Egnazia in Apulia, southern Italy. “We must give importance to, guarantee and protect human control over the selection process of artificial intelligence programs,” the 87-year-old demanded. Because: “Human dignity is at stake.”
Advertisement
Even “in a drama such as armed conflict” it is “urgently necessary” to reconsider the development and use of devices such as “lethal autonomous weapons”, the head of the Roman Catholic Church warned the gathered heads of state and government of important industrialized countries. He explicitly advocated a moratorium on the use of such technology. Until it is ready for decision, “an active and concrete commitment to introduce greater and more significant human control” is needed.
This was the first time in the history of the G7 institution that a pope was a guest. Francis had no doubt that the emergence of AI “represents a truly cognitive-industrial revolution.” It will “contribute to the creation of a new social order” that is “characterized by complex, epochal changes.” For example, the key technology could lead to “the democratization of access to knowledge, the rapid progress of scientific research and the possibility of entrusting difficult tasks to machines.” However, he devoted most of his remarks to the risks of AI. According to him, it could also lead to “greater injustice between advanced and developing countries, between dominant and oppressed social classes.” The “culture of encounter” is in danger of being replaced by a culture of “throwbacks.”
Criticism of AI as an aid to judicial decision-making
For the pope, rapid technological progress makes artificial intelligence “a fascinating and at the same time scary tool. It requires a reflection that does justice to the situation.” Francis himself has become part of AI meme culture with a photo of himself wearing a rapper-style white designer padded jacket.
Demand for quick digital pact 2.0 for schoolsIn His lectures He now pointed to a major difference: “It must always be kept in mind that machines can make algorithmic choices in some forms and with these new tools.” This approach is based “either on precisely defined criteria or on statistical inferences.” Human beings, on the other hand, not only choose, “but are also capable of making decisions in their hearts.”
AI is therefore designed to “solve specific problems”, Pope explained. However, for users, the temptation is often “irresistible to draw general, even anthropological conclusions from the specific solutions they propose.” An example is the use of programs to help judges decide whether to place convicted criminals under house arrest. Here, AI is used to estimate the likelihood of reoffending based on predefined categories. The relevant systems would have access to private data categories such as ethnic origin, level of education or credit lines. There is not only a great risk of “letting a machine make the final decision about a person’s fate”, but also of “the underlying context of predefined assessments”, i.e. the AI adopting prejudices or stereotypes from the training data.
“Generative” AI is not creative, but “amplifying”
Francis thinks little of knowledge systems and bots like ChatGPT from OpenAI or Gemini from Google. “Many of us are impressed by the applications that are readily available on the Internet to write text or create images on a topic,” he said. “Students especially like it because they use it infrequently when they have to work,” he couldn’t help but take a dig at it. However, students forget that, strictly speaking, so-called generative AI is not really “creative”. What it is actually doing is mining big data for information and packaging it in the style it wants. The more often such a system encounters a term or hypothesis being repeated, the more likely it is to consider it valid and valid.
Pope pointed out that such large language models are “reinforcing” in the sense that they rearrange existing content and help consolidate it. However, they did not check whether there were errors or biases. This not only risks “legitimizing fake news and reinforcing the gains of a dominant culture”, but also undermines the educational process at its core. There is no longer room for the authentic reflection that is really needed.
Francis identified the main weakness in the implementation and development of these systems as “the lack of an ethos linked to the notion of the value and dignity of the human person.” For AI to become a tool “for goodness and the creation of a better future,” it “must always aim at the good of every single person.” The Vatican and tech companies signed an agreement with the Italian government in 2020 Call for ethical orientation of algorithmsAI should not reinforce the “technological paradigm,” the Pope said, according to which “the view of the world is limited to realities expressed in numbers and prefabricated categories.”
(d)
EU set to impose DMA fine against Apple over App Store
