Elon Musk makes a mistake when asking for a break from AI, experts say – 03/31/2023 – Tech

Elon Musk makes a mistake when asking for a break from AI, experts say – 03/31/2023 – Tech


The letter from technology experts and entrepreneurs, including Elon Musk, questioning the state of artificial intelligence (AI) even has good intentions, but the request to stop technology development is “very problematic”, argues Google engineer Luiz André Barroso , former vice president of the company.

“I agree with the intent behind this. Obviously [a IA] it is very important for society, it has to be managed very carefully. [Mas] The solution proposed by this letter is very problematic. The benefit capacity of these technologies is enormous, and we are not incapable of managing the benefits ethically and correctly”, said Barroso this Friday (31).

This opinion was echoed by colleagues who shared with him a lecture on the future of AI at the Brazil Conference, an event organized by students from Harvard and MIT (Massachusetts Institute of Technology) universities, in Cambridge, in the Boston area, in the USA. The conversation was moderated by AI expert Carmem Domingues.

Elon Musk, historian Yuval Noah Harari and hundreds of experts released on Wednesday (29) a letter in which they call for a six-month pause in research on artificial intelligence more powerful than GPT-4, the OpenAI model launched this month.

They argue that the uncontrolled development of artificial intelligence poses “great risks to humanity” and call for a moratorium until safety systems are established with new regulatory authorities; surveillance of AI systems; techniques that help to distinguish between the real and the artificial; institutions capable of coping with the “economic and political disruption (especially to democracy) that AI will cause”.

“I find the letter very unfortunate and sad,” said Marcus Fontoura, Stone’s CTO. “AI has existed since 1970 and we always wanted the computer to have a form of logical reasoning [raciocínio lógico] that we could interact, it’s nothing new. What is new is new data. There is governance over data, regulation, we are very far from a science fiction scenario in which robots will dominate the world. The more investment in this technology, the more applications will emerge, in education, sustainability. Anything detracting from that is very sad,” he said.

For Harvard professor of electrical engineering Flavio Calmon, “technology has to be developed responsibly”, but it is not credible to think about interrupting the use of machine learning algorithms, research in mathematics and data collection. “Certainly the letter will have no effect on the development of these algorithms,” he said.

According to Calmom, ChatGPT became so popular because it arrived at a perfect time when there is a huge amount of data collected with an unprecedented computational capacity.

Barroso, from Google, played with the limitations of the artificial intelligence tool, which can sometimes give meaningless answers to certain problems. “Because they are language models, they are much more fluent than competent. The ability of these bots to answer questions is almost inversely proportional to the authority they have on the subject,” he said.



Source link