Google Opens Bard, Its ChatGPT Competitor, for Public Testing

pcmag.com:

Ahead of the public release of its homegrown Bard AI search tool, Google is inviting users in the US and UK to try it out themselves.

Anyone can join the waitlist to be a tester at bard.google.com(Opens in a new window), and Google will email you when you have access. It's an "early experiment," Google says(Opens in a new window) in a blog post authored by Sissie Hsiao, product VP, and Eli Collins, research VP. The public beta will focus on improving the tool's accuracy by giving it more data to crunch.

The beta experience sounds more like a choose-your-own-adventure book than a typical search inquiry. "When using Bard, you'll often get the choice of a few different drafts of its response so you can pick the best starting point for you," Google says. "You can continue to collaborate with Bard from there, asking follow-up questions. And if you want to see an alternative, you can always have Bard try again."

Google announced Bard in early February—just over two months after OpenAI launched ChatGPT, and one month after Google issued an internal code red on the potential for ChatGPT to change the field of search and overtake Google as top dog. At the time, internal employees were asked to test out Bard, CNBC reports(Opens in a new window).

Unfortunately, the announcement included a screenshot of Bard inaccurately answering a question about the James Webb Space Telescope. Given the volume of search queries on Google, this stirred up concerns about the company moving too fast to incorporate large language models (LLMs) and misinforming the public.

"While LLMs are an exciting technology, they’re not without their faults," Google says. "For instance, because they learn from a wide range of information that reflects real-world