The risks associated with artificial intelligence are well documented: privacy violations, theft of intellectual property or the great replacement of the human worker, to name a few. But Pilar Manchón, an AI expert at Google, highlights another: “human stupidity.” For Manchón, this is the greatest risk that must be taken into account when regulating this technology, as she indicated this Tuesday during the panel ‘The state of generative AI in the company: Now the next one decides’ held within the framework of the forum Latin America, the United States and Spain in the global economyorganized in New York by EL PAÍS and the Spain-United States Chamber of Commerce.
“We know that we will go so far and so fast because we are in control. So it is up to us to make sure that we take the necessary steps and precautions to use AI in the most constructive, fair and appropriate way,” explained Manchón, Senior Director of Engineering, AI Research Strategy at Google, during her speech at the event sponsored by DLA Piper, Iberia, Inditex, Indra, NTT Data and Total Protect in collaboration with the Organization of Ibero-American States (OEI). The expert also stressed that “AI is just a tool, it is a means to an end.” Therefore, it is humans who decide how they use it, for good or bad. “In this area, I would say that we are our best allies, and also our best and most dangerous enemies,” she added.
To address this and other risks, Manchón believes that a common regulatory framework is needed that combines those already in place in Europe and the United States, among others. “Different societies, in Europe, the United States or Asia, will have different interpretations of how much risk we want to take and how much sensitivity there is about certain data, certain applications, certain things. So we have to find a way to make compatible the different legal frameworks that are being developed in Europe, the United States and other parts of the world. And we have to reach an understanding in which mobility and transience enrich our world,” he said.
It’s about finding “the balance between risk and value,” said Domingo Gonzalez, Nestlé’s director of IT innovation for North America. “And depending on the risk, there are some safeguards and some compliances that need to be met,” he said. Some of those safeguards can be addressed through education: “Whenever we deploy generative AI tools at Nestlé, we spend a lot of time on training, teaching people how to use the tool, how to talk to it, how to get value from it.” “Thinking at a country level, we have to teach citizens the same thing,” he added.
“I think we all know that AI is too important not to regulate it, but it is also important not to regulate it excessively,” Manchón warned, agreeing on the need to find the “right balance” that González referred to. “It is also important to make sure that we are not regulating the technology, but the applications of the technology, because technology is not good or bad, what matters is how it is used, and it is up to us how we use it,” Manchón added.
For his part, Mariano Jabonero, Secretary General of the OEI-Organization of Ibero-American States, highlighted during a second panel focused on artificial intelligence that the regulation of this technology must also prioritize the protection of intellectual property. “We have created a specific chair on the subject because if there was a strong plundering in the region with physical heritage, which is historical, now we find ourselves with a plundering of the work of creatives who live off of that, which is completely unprotected. We must pressure governments in that direction because it is part of the wealth of the region,” he warned.
AI for all
The second panel on AI also addressed the possible uses of AI in sectors such as education, science and culture. But both Carme Artigas, co-chair of the United Nations advisory body on AI, and Jabonero, stressed the importance of the tools reaching all countries, not just the rich ones.
“We need to compensate for the huge differences between the global north and south,” Artigas said. “Of the 100 supercomputers in the world, there is not a single one in the global south. We are in a geopolitical battle for the control of raw materials, of chips… We do not want a new techno-colonialism, we want Latin America and Africa to be able to generate their own solutions with the tools that the north can provide,” he added.
#Googles #Pilar #Manchón #biggest #risk #face #human #stupidity