Google's AI Chatbot Bard Falls Behind Microsoft's ChatGPT</h3>
On February 6, a promotional video was shared on Twitter showcasing Bard, the chatbot rival to Microsoft's ChatGPT. In the video, Bard gives an incorrect answer when asked about the discoveries from the James Webb Space Telescope. This error was noticed by astrophysicist Grant Tremblay and reported by Reuters and the New Scientist.
Bard vs Bing
On February 6, Alphabet CEO Sundar Pichai announced the company would be opening Bard to trusted testers with the intention of releasing it to the public shortly after. On February 8, Google was supposed to host a launch event for the experimental chatbot. However, Microsoft stole the show with a surprise event the day before where they unveiled a new version of their search engine Bing, developed with OpenAI, the company behind ChatGPT. This new Bing is running on a more powerful language model than ChatGPT, customized specifically for search.
Google's Launch Event Falls Short
Reports of Bard's error overshadowed Google's event, which failed to live up to Microsoft's demonstration. Google's stock fell by 9% by market close on February 8, while Microsoft's share price rose by 3%. According to Mashable tech reporter Cecily Mauran, who was at the event, it felt like a last-minute addition that didn't meet attendees' expectations.
Google's Rush to Release AI Chatbot
Once a leader in AI, Google has been leapfrogged by OpenAI in recent years. The growing popularity of ChatGPT led Google to declare the development of its own chatbot as a "code red" project. This, along with Microsoft pumping billions more into OpenAI, put pressure on Sundar Pichai's team to prioritize the development of a response to ChatGPT.
Despite this pressure, errors like the one in the Bard ad are far from uncommon and can be found in several similar models. However, in the rush to get ahead, Google can't ignore Bard's shortcomings.
An AI "Dance of Giants"
ChatGPT has reached 100 million users as of February 1. Google, on the other hand, doesn't need to gain users, it just needs to roll out to its existing search box. If and when Bard is fully rolled out, over 1.1 billion users will have access to these experimental chatbots, making it, as Jim Fan, an AI research scientist at Nvidia, puts it, "a dance of giants."
A Google spokesperson has stated, "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."
Technomancer is a science and tech enthusiast who enjoys writing about software and AI and other tech topics.