One Wrong Answer From Bard AI Results To Google Loss Of $100 Billion

Google's Bard AI, a new experimental conversational tool, and rival to Microsoft's ChatGPT, has resulted in significant losses for the tech giant. On Wednesday, Alphabet, Google's parent company, saw its stock drop 7.7% and lose over $100 billion in market value after its new AI gave an incorrect answer to a query.

The Internet giant introduced Bard with high hopes and even organized a live stream event to showcase its capabilities. However, when Bard was asked to explain new discoveries from the James Webb Space Telescope to a 9-year-old, the AI provided two correct answers but made a mistake while answering the third one. 

"What new discoveries from the James Webb Space Telescope can I tell my 9-year-old about?" The newly unveiled Google AI responded that "JWST (James Web Space Telescope) took the very first pictures of a planet outside of our very own solar system."

However, Grant Tremblay, an astrophysicist from the Harvard-Smithsonian Center for Astrophysics, didn't agree with the answer and then pointed out the error in a tweet.

"I'm sure Bard AI will be impressive, but for the record: JWST did not take "the very first image of a planet outside our solar system". the first image was instead done by Chauvin et al. (2004) with the VLT/NACO using adaptive optics," the tweet read.

Not to be a ~well, actually~ jerk, and I'm sure Bard will be impressive, but for the record: JWST did not take "the very first image of a planet outside our solar system".

the first image was instead done by Chauvin et al. (2004) with the VLT/NACO using adaptive optics. https://t.co/bSBb5TOeUW pic.twitter.com/KnrZ1SSz7h

— Grant Tremblay (@astrogrant) February 7, 2023

According to NASA, the first picture taken of a planet outside was done using the Very Large Telescope in 2004, which is almost 19 years before NASA's Webb telescope.

A Google spokesperson acknowledged the error and emphasized the importance of thorough testing, which the company plans to kick off with its Trusted Tester program.

"This highlights the importance of a rigorous testing process, something that we're kicking off this week with our Trusted Tester program. We'll combine external feedback with our internal testing to make sure Bard's responses meet a high bar for quality, safety, and groundedness in real-world information," A Google spokesperson told New Scientist.

In another tweet, Tremblay also noted how a Google search using the regular search engine gives the correct information.

This error highlights the limitations of AI models, which are based on statistical analysis and rely on plausible answers rather than accurate ones. The news of Google's losses also comes as Microsoft has recently launched its own AI results service for its Bing search engine, and Chinese search engine Baidu has announced similar plans.