Experts worry that artificial intelligence will fall “into the wrong hands” / freepik
Artificial intelligence (AI) could cause “real damage” if it falls into the wrong hands, Microsoft chief economist Michael Schwarz said yesterday, calling for “regulation” on the matter.
“I am confident that AI will end up being used by bad actors, and that, in that case, it will cause real damage,” Schwarz warned when speaking at a panel at the World Economic Forum in the Swiss city of Geneva.
Faced with its malicious use, for example, in electoral elections, the economist opined that the technology “clearly” should be regulated, although he asked that the authorities be cautious and wait until it causes the first negative effects.
“The principle should be that the benefits of regulation to our society should be more important than the cost it causes,” he argued in statements released by the Bloomberg agency.
Following the explosion in the use of technology due to the debut of the ChatGPT program, various regulatory authorities are pressing the sector to implement limits. Currently, seven countries have banned the use of said chatbot, including Italy, while other countries such as the United Kingdom have already announced plans to regulate the matter. In the Italian case, the regulatory authorities asked the developer of the program (OpenAI) to stop processing the data of Italian users, while they investigate a possible breach of European privacy regulations.
OpenAI recently received a millionaire investment from Microsoft of US$ 10,000 million. As a result, the multinational has already implemented the chatbot in its Bing search engine. According to Schwarz, Microsoft is working on mitigating the potential harm from artificial intelligence tools.
The economist affirmed that technology will allow people to be more productive, “producing more with less work”, and stressed that it will “revolutionize” the way in which companies operate.
“AI is not going to change anything in the short term, but it will change everything in the long run, as has happened with every technology in the past,” he explained.
Impact on employment
According to a report by the World Economic Forum published this week based on business surveys, by 2027 42% of business tasks will be automated (compared to 34% today), including 35% of reasoning and decision-making tasks. decision, and 65% in the case of data and information processing.
“Reasoning, coordination and communication, all traits in which humans have a comparative advantage, will become more automatable in the future. AI, a potential algorithmic replacement, is expected to be adopted by nearly 75% of companies surveyed,” the report states.
The report precisely indicates that the areas specialized in AI and “machine learning” will be the jobs with the greatest growth potential in the coming years.
Meanwhile, according to a Goldman Sachs study, some 300 million jobs could be automated.
One of the first companies that already gave concrete figures in this regard was IBM. The CEO of the multinational, Arvind Krishna, anticipated this week that “easily 30% of customer service roles could be replaced by automation and AI in a period of five years.” This would translate into 7,800 lost jobs out of the nearly 26,000 who perform such tasks at IBM.
In this framework, Krishna indicated that IBM, for now, will pause hiring in roles in which it is believed that AI can replace them in the coming years.
The director of IBM exemplified that some of the replaceable roles will be those linked to some functions of human resources and administrative tasks, and not those of software development, the core of the firm that currently has almost 260,000 workers.
As it will be remembered, the UN warned a month ago about the dangers of using Artificial Intelligence and asked companies to be responsible after seeing how ChatGPT managed to alter images, replicate voices and create texts.
At the same time, technological leaders such as Steve Wozniak, Jaan Tallinn or Elon Musk joined a petition in March that seeks to temporarily stop large experiments with AI due to the risks it can entail for society and proposed implementing security protocols.
In Spain, the Spanish Agency for Data Protection reported that it had initiated ex officio prior investigation proceedings against the OpenAI company, owner of the ChatGPT service, for a possible breach of the regulations.
Studies also warn that some 300 million jobs could be automated
