Google Announces Its Own Chatbot, Bard, to Answer ChatGPT

Google's chief executive said the company will open a chatbot called Bard to testers now and to the public in the coming weeks, built on a program Google has had for two years but never released.

Sundar Pichai, Google's chief executive, announced in a blog post today that the company is releasing a chatbot called Bard. It opens to a group of "trusted testers" immediately and to the public in the coming weeks. It is built on LaMDA, the conversational program Google has had since 2021 — the same one an engineer claimed last summer had come to life.

Google invented most of the technology behind ChatGPT. The design that makes these programs work was published by Google researchers in 2017, and Google has had chatbots at least as capable as OpenAI's for years. It didn't release them. The company's stated reason was caution: a chatbot that states falsehoods confidently is a liability for a company whose whole business is being the place you go to find things out. Its unstated reason, as many inside and outside the company have said, is that a search engine that answers your question is a search engine that shows you fewer ads.

ChatGPT changed the calculation. Reports in December described a "code red" inside Google, with the founders, Larry Page and Sergey Brin, called back to review the company's AI plans. Microsoft is expected to show its own chatbot, built with OpenAI, tomorrow.

Pichai's post is careful. Bard will be a "lightweight" version of the program at first, to serve more people cheaply, and Google will "combine external feedback with our own internal testing" to make sure its answers meet "a high bar for quality, safety and groundedness." Translated: it will be wrong sometimes, and Google knows it.