The head of the Google search engine warned in an interview published today (Saturday) about the dangers of artificial intelligence in chatbots. The warning comes as Google struggles to compete with the popular ChatGPT app.
“The artificial intelligence we are talking about can sometimes produce what are known as hallucinations,” Prabhakar Raghavan, Google Vice President and Head of Search Engine, told Welt am Sonntag.
By hallucination, Raghavan explained, he means a kind of artificial intelligence that speaks in a way that gives convincing answers but is completely made up. He added that you should first try to reduce hallucinations.
Raghavan's warning comes amid a series of artificial intelligence projects at Microsoft and rival Google. Interest in AI capable of generating scripts has grown since the launch of the GBT chatbot on the web in late November.
Microsoft, which already has a close partnership with OpenAI, the developer of the GBT chatbot, wants to capitalize on this interest by announcing at a special event last Tuesday a new version of its Bing search engine, based on the updated version from the same day in Synthetic. Intelligence technology that powers GPT Chat.
The launch of ChatGBT is said to have set off alarm bells within the search giant as longtime co-founders Larry Page and Sergey Brin have been brought in to help with what could be a threat to the company's biggest revenue engine.
Google announced its chatbot, called Bard, on Monday, but the bot provided false information that cost Alphabet about $100 billion in market value as of Wednesday's close, Reuters reported.
For this, Google is still testing Cool and has not indicated when it will be available to the public. "Of course we feel the urgency, but we also feel the responsibility," Raghavan said. We certainly do not want to mislead the public. "