Google AI chatbot Bard gives wrong answer, sending shares plummeting

Amid the rise of OpenAI’s ChatGPT, Google is trying to get in on the action with its own chatbot, Bard, but a factual error from the AI tool just cost the tech giant approximately US$100 billion.

In a promotional clip shared by Google on Monday, the generative AI chatbot is asked, “What new discoveries from the James Webb Space Telescope can I tell my nine-year-old about?” Bard answers with several bullet points, one of which reads, “JWST took the very first pictures of a planet outside of our own solar system.”

Experts online were quick to point out the error. Grant Tremblay, an astrophysicist at the US Center for Astrophysics, replied to Google on Twitter, writing, “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.'”

According to NASA, the first-ever image of a planet outside our own solar system was taken in 2004 by the European Southern Observatory’s Very Large Telescope (VLT).

Shares for Alphabet, Google’s parent company, subsequently fell 7.7 per cent on Wednesday, resulting in over US$100 billion in lost market value, as per Bloomberg. The change was the most severe stock drop the company has seen in over three months, when an Alphabet earnings report in October 2022 disclosed a slowdown in digital ad revenue.

In contrast, Microsoft’s stock leapt 4.2 per cent on Tuesday after the tech giant announced ChatGPT would be used to revamp and power the search engine Bing.

A Google spokesperson told CNN in a statement Wednesday that Bard’s error “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 own internal testing to make sure Bard’s responses meet a high bar for quality, safety and groundedness in real-world information,” the spokesperson said.

Though Bard has yet been released to the general public, the company said the chatbot would be available in the “coming weeks.”

Bing, even with the implementation of ChatGPT, is likely still a long way away from posing a serious threat to Google, which holds a roughly 90 per cent share of the internet market.

“It’s not like this is the end of the world for Google and Microsoft is going to eat its lunch in search,” CFRA analyst Angelo Zino said. “But ChatGPT is showing there is possible threat and that is causing a lot of fear.”

Google’s Bard is very similar to ChatGPT in that users put in a question, request or give a prompt to receive a human-like response.

ChatGPT, which was released to the public last November, can be anything the user makes of it — it can take on the role of a chef and provide recipes, make business plans for marketers, create press releases for public relations specialists or give advice like a therapist.

Like Microsoft, Google plans to embed AI tools in its Google Search, which makes up a large portion of the company’s revenue.

The biggest difference between ChatGPT and Bard is the latter’s ability to include recent events in responses. Bard will draw on information from the internet, while ChatGPT has access to data from various sources (up to date as of 2021).

Both tools are built on a large language model, but experts have long worried the AI systems could lead to the potential spread of misinformation. However, with further development, AI experts believe the tools can be taught to discern between correct and incorrect information.

“We will see an era of incredible experimentation, and probably fairly rapidly convergence to the tools that are actually meeting the needs of the people who are meant to use them,” Mark Daley, chief digital information officer at Western University, previously told Global News.

— With files from Global News’ Aaron D’Andrea and The Associated Press