Following its launch on Monday, the new AI search assistant from Google was asked: “what new discoveries from the James Webb Space Telescope can I tell my 9 year old about?”, to which one response was “JWST took the very first pictures of a planet outside of our own solar system”, reported the Telegraph.
Bard’s answer was incorrect — the first image taken of an exoplanet was by the Very Large Telescope array in Chile in 2004, while NASA’s James Webb Space Telescope was made operational in December 2021.
One astronomer, according to the Financial Times, shared the view that the Bard chatbot could have misinterpreted “ambiguous Nasa press releases that underplayed past history”.
Two more responses, as revealed by a promotional GIF animation shared via Google’s Twitter account, were:
- “In 2023, The JWST spotted a number of galaxies nicknamed “green peas”;
- “The telescope captured images of galaxies that are over 13 billion years old”.
The demonstration glitch, which was discovered prior to a presentation, led to Google’s parent company Alphabet losing $100bn in market value on Wednesday, and demonstrated the risks of shared misinformation that generative AI systems can cause.
>See also: The ethical implications of chatbots
In response, a Google spokesperson said: “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 own internal testing to make sure Bard’s responses meet a high bar for quality, safety and groundedness in real-world information.”
Additionally, Google senior vice-president Prabhakar Raghavan announced search updates at an event in Paris this week, including new image-based search and interactive 3D models of cities such as London and Los Angeles.
Meanwhile, Microsoft‘s share price has grown 6 per cent over the past week, with analysts claiming that its launch of ChatGPT for Bing and Edge — upon entering a new partnership phase with OpenAI — was the main factor.
It is currently unknown when Google’s Bard AI chatbot will be made publicly available, as is how exactly the algorithm is being trained to generate answers and search results.
Related:
How AI could be a game-changer for data privacy — AI offers multiple benefits to businesses, but it also poses data privacy risks.
Unlocking the ‘black box’ of education data — Exploring how the education sector can maximise the value of data.
