Microsoft has integrated OpenAI tools such as ChatGPT into its Bing search engine as part of a massive investment in AI. By: JASON REDMOND/AFP < p>Artificial intelligence (AI) will «do real damage» when it falls into the hands of «malicious people,» Microsoft CEO warned.
Michael Schwartz, chief economist at a US technology company, told reporters World Economic Forum (WEF): «I'm sure that AI will be used by bad actors, and yes, it will cause real damage.»
«It can cause a lot of damage in the hands of spammers associated with elections and so on» , he said in comments published by Bloomberg.
His warning echoes that of Geoffrey Hinton, the “godfather of AI,” who is leaving Google this week.
Mr. Hinton said in a newspaper interview: «It's hard to see how you can prevent attackers from using it for bad purposes.»
Mr. Schwartz's comments came after the bosses of Microsoft, Google and ChatGPT creator OpenAI were called to the White House to meet with U.S. Vice President Kamala Harris on Thursday.
The meeting was convened so officials could have a «candid discussion» with technology executives about the risks of future AI technologies.< /p>
Scientists from Microsoft and OpenAI said in March they saw «sparks» of human intelligence in the latter's latest creation during lab tests.
Computer scientist Professor Geoffrey Hinton, known as the «godfather of AI,» resigned from Google this week over concerns about the technology. Photo: Julian Simmonds.
This so-called artificial general intelligence could herald the creation of autonomous robots. capable of making their own decisions, completely independent of human input.
OpenAI's November release of ChatGPT took the world by storm as the AI chatbot exceeded all previous expectations of this technology.
Its uniqueness the advantage is that it has been trained on petabytes of data from all over the internet, which means that users can ask it questions on almost any topic and get convincing-sounding answers.
Experts warn that AI-powered chatbots, however, are capable of «hallucinating» answers that could be completely wrong.
«The kind of AI we're talking about right now can sometimes lead to we call hallucinations,” Prabhakar Raghavan, head of Google Search, reported a German newspaper in February.
About US$120bn (£96bn) was struck off Google’s market value later that month after , as its ChatGPT competitor, Bard, gave a convincing but incorrect answer to the question. The answer was then reproduced in promotional materials for the launch of the chatbot.
Bard gave an inaccurate answer to the question: «What new discoveries of the James Webb Space Telescope can I tell my 9-year-old child about?»< /p>
This is a false assumption that the telescope was used for the very first discoveries of planets outside the solar system.
The response caused the biggest one-day drop in Google's share price in five months.
Свежие комментарии