Google’s Bard AI chatbot, launched as competition for ChatGPT, is reportedly unable to correctly answer SAT questions and has issues with math and writing.
Fortune reports that as Google opens its AI language model, Bard, to public testing, it has become apparent that the system has a long way to go before achieving academic success. Fortune recently put Bard through its paces ahead of the upcoming SATs, revealing that the AI struggled with the majority of math questions and had difficulties acing writing and language tests.
Sundar Pichai, the CEO of Alphabet, is still upbeat about Bard’s prospects despite these setbacks. In a company-wide memo, Pichai wrote, “As more people start to use Bard and test its capabilities, they’ll surprise us. Things will go wrong.” He emphasized that Bard is still in its early stages and that users should expect a learning curve.
Even when multiple-choice answers were available, Bard failed to correctly answer 50 percent to 75 percent of the SAT math questions during the testing process. But after a few days of testing, the AI got better at reading and writing exercises, and its accuracy increased.
A Google spokesperson reiterated Pichai’s message, saying, “Bard is experimental, and some of the responses may be inaccurate, so double-check information in Bard’s responses. With your feedback, Bard is getting better every day.”
This isn’t the first time Google has had trouble with the Bard chatbot. Bard’s public introduction turned into a disaster for Google’s stock price.
Reuters reports that this week, Alphabet, the parent company of Google, saw a $100 billion decline in market value as a result of Bard, the company’s new chatbot, providing inaccurate information in a promotional video. Because of this failure and the lack of information regarding Bard’s integration into Google’s primary search function, there are worries that Google is falling behind its competitor Microsoft. The stock fell as much as 9 percent during Wednesday trading.
The launch of Google’s new chatbot on Monday was anticipated to simplify complex subjects. But just before its livestreamed presentation on Wednesday, Reuters called attention to a mistake in the company’s promotional video — the supposedly all-knowing AI made a basic mistake about space exploration.
Read more at Fortune here.
Lucas Nolan is a reporter for Breitbart News covering issues of free speech and online censorship. Follow him on Twitter @LucasNolan
Read the full article here
Discussion about this post