The Bard AI bot error cost Google $100 billion in stock value

Posted by

Google is looking for new arguments to make its case as the leader in AI.

And the internet behemoth appears to be guessing wrongly thus far.

To demonstrate the capabilities of its new AI bot, an advertisement provided an inaccurate response to a test question.

As of Wednesday, the market value of Alphabet, the company’s parent company, has dropped by $100bn (£82bn) due to a drop of more than 7% in share price.

On Monday, a promotional tweet for the bot, dubbed “Bard,” was sent to Twitter, in which it was asked what it would tell a nine-year-old about discoveries made with the James Webb Space Telescope.

Astronomers on Twitter were eager to point out the error in the statement, which stated that the telescope had taken the first images of a planet outside the solar system. In reality, the European Very Large Telescope accomplished this feat in 2004.

“Why didn’t you double-check the details before citing that example?” Newcastle University postdoctoral researcher Chris Harrison replied to the tweet.

The presentation the corporation provided to investors about its aspirations to implement AI in its goods was similarly met with a lukewarm reception.

  • Google releases an alternative to ChatGPT called Bard.
  • Microsoft releases a new Bing enhanced with ChatGPT technology.

Similar to Google but more effective than Google It’s a chatbot arms race thanks to ChatGPT
Since OpenAI, supported by Microsoft, released updated ChatGPT software late last year, Google has been under intense scrutiny. Its ability to ace MBA quizzes, write catchy song lyrics, and solve a wide variety of other problems led to its rapid spread online.

Microsoft announced this week that the next iteration of its Microsoft search engine, which has lagged behind Google for years, will employ an upgraded version of the ChatGPT technology.

Some have warned that the hasty release of AI technology increases the likelihood of errors or biased findings and can even lead to issues of plagiarism, despite the fact that investors have applauded the push for AI.

A Google representative claimed that the blunder demonstrated the need for “a rigorous testing procedure,” which the company is beginning this week with the launch of its Trusted Tester program.

“We’ll combine external evaluation with our own inside testing to ensure sure Bard’s answers satisfy a high bar for quality, safety, and roundedness in real-world knowledge,” they stated.

In the midst of layoffs at a number of top tech titans last month, Google’s parent company Alphabet lost 12,000 employees or nearly 6% of its worldwide staff.

 

Leave a Reply

Your email address will not be published. Required fields are marked *