Was the public release too hasty... Google challenges MS amid AI Bard error controversy

[Asia Economy New York=Special Correspondent Joselgina] Google has launched a counterattack against Microsoft (MS), which equipped its search engine with the AI chatbot 'ChatGPT.' Google now supports more direct and longer text responses to complex questions and has integrated AI into Google Maps and Translate. However, concerns about accuracy are spreading as Google's ambitious 'AI Bard,' introduced as a rival to ChatGPT, was found to have provided incorrect answers.

According to major foreign media, on the 8th (local time), Google revealed detailed information on how AI technology is applied to its new search features, Google Maps, and Translate at an AI event held in Paris, France. The announcement came just one day after MS unveiled a new version of 'Bing' equipped with an AI chatbot. This marks Google's counterattack in the big tech AI technology battle over the search market.

First, Google explained that it will provide more direct and longer text answers to complex questions, rather than the previous responses such as short answers, snippets, or external site links.

Prabhakar Raghavan, Google Senior Vice President, said, "We are helping people understand information in a completely new way beyond the traditional concept of search," and explained that generative AI like the recent ChatGPT craze "will help organize complex information and diverse perspectives immediately." The company confirmed that the launch timing will be possible when Google is confident in the quality of AI-provided answers.

Additionally, Google announced that Google Maps users will be able to explore information extracted by AI beyond just 2D images. By applying AI technology to the existing immersive view, users can vividly examine restaurant interiors and landmarks of famous cities. Along with this, improvements were confirmed in contextual translation for five languages: English, French, German, Spanish, and Japanese. Furthermore, the multi-search feature, which allows adding search terms to image searches for more accurate answers, will be launched worldwide.

Google's announcement came amid the recent surge of ChatGPT and the full-scale integration of search engines and AI chatbots. Sundar Pichai, CEO of Google's parent company Alphabet, confirmed on the 6th that the AI chatbot-based 'Bard AI' will be launched soon. He said the new AI feature will "filter complex information and perspectives into an easily digestible format" and that they will "continue to introduce the best Google AI for a better life."

The event also briefly introduced Bard. Bard is based on 'LaMDA,' a large language model trained with 137 billion parameters. However, major foreign media reported that Bard provided some incorrect answers in the promotional video released during Google's announcement.

When asked, "How would you explain the new discoveries of the James Webb Space Telescope to a 9-year-old child?" Bard included the answer that it was "used for the first time to photograph exoplanets outside the solar system," which was pointed out as a factual error. According to NASA, the first such photograph was taken by the VLT (Very Large Telescope) installed by an astronomy research team led by Gael Chauvin of the European Southern Observatory. Concerns about Bard's reliability have emerged, and Alphabet's stock price was declining in the New York stock market that day.

Following MS and Google, Chinese big tech companies are also entering the generative AI competition like ChatGPT, intensifying the big tech AI technology war. After Baidu announced plans to launch the AI chatbot 'Ernie Bot' next month, China's largest e-commerce company Alibaba also confirmed it is internally testing generative AI technology similar to ChatGPT.