Google’s AI chatbot, Bard, gave a wrong answer in its promotional video, underscoring the integrity risks that AI systems pose to corporations. Shares of the company tumbled 7.8 percent on the Nasdaq exchange in the regular trading hours on February 8 and were essentially flat in the after-hours trading.
The error was caught hours before Bard’s launch event in Paris. The chatbot delivers an incorrect answer to a prompt in the video showing Bard in action, reports MSN.
Google Inc’s new experimental AI service, ‘Bard,’ which was developed to answer Microsoft’s ChatGPT, delivered inaccurate information in its promotional video.
Reuters caught the error a few hours before Bard’s launch event in Paris. In the video showing Bard in action, the chatbot responds to a question about what to tell a 9-year-old about discoveries from the James Webb Space Telescope (JWST), saying that the JWST was used to take pictures of a planet outside the Milky Way. This is not true.
The chatbot, which aims to simplify complex topics, was built to compete with Microsoft’s ground-breaking AI chatbot, ChatGPT. However, Google seems to have come up with a rival too quickly.
Generally speaking, there still is a long way to go for AI systems to be incorporated into business operations and replace jobs. The risks of plagiarism, skewed results, and inaccurate solutions remain.
Related Stories:
Google Opens Public Access To Bard, Its AI ChatGPT Rival at Bard.google.com
Google to introduce AI chatbot service ‘Bard’ to rival Microsoft’s ChatGPT
Google I/O 2023: Google Bard arrives in India: Features, benefits and challenges
Why Google Bard is not accessible in EU and Canada
Google Bard’s Leaked Demo Unveils Assistant’s New Features
Google I/O 2023 Opens with a Stunning AI Music Performance by Bard and Dan Deacon
Google reveals new AI tools and features for Workspace, Search, and Bard
Bard is now available in over 40 new languages and 27 countries in EU
