Google's Bard chatbot will launch in the United Kingdom and United States on Tuesday, completing the company's rush to release a competitor to Bing Chat and ChatGPT.

It is viewed as a life-or-death situation for the company, whose profitable web search service risks being outcompeted by artificial intelligence (AI) chatbots, despite the fact that these chatbots presently struggle to consistently return accurate and useful results.

ALSO READ: What is Google Bard?

Google's Jack Krawczyk, the product lead for Bard, referred to the service as a "experiment" and said the company was "extremely thrilled to observe how people are using this product to enhance their creativity. It aids in the acceleration of their ideas and in the stimulation of their inquiry.

"We believe we've reached the end of the testing portion of this experiment," Krawczyk continued. "Now, we'd like to begin rolling it out gradually. We are at the very beginning of the transition from research to reality, and we are about to endure a long arc of technological development."

Like its competitors, Bard allows users to type queries and requests in standard English and is capable of providing creative responses to challenging inquiries and engaging in back-and-forth conversations. On Tuesday, users can sign up for access on the company's website via a waiting list.

In a live demonstration, Krawczyk demonstrated how Bard assisted him in coming up with ideas for a birthday party for his son that incorporated his two passions, rabbits and gymnastics.

Using one of Bard's unique features, the ability to easily click between three answers to the same question, he selected a suggestion that was more focused on the "gymnastics" part than the "rabbit" part. He then instructed the chatbot to generate an email invitation and insert the address of the actual venue obtained through a search.

However, the unanticipated continuation was less impressive. In response to a request for vacation ideas for two children in Tokyo, Bard initially suggested a visit to the Tsukiji fish market, but neglected to mention that a significant portion of the market relocated to a new location in 2018. The chatbot acknowledged this in a subsequent query by providing footnotes to the correct information.

ALSO READ: The top AI chatbots: ChatGPT and additional notable alternatives

"The answers aren't going to be perfect along the way," Krawczyk said, adding that for certain queries, a traditional Google search will continue to be the best option. "There is a great deal of written content on the internet where you can still conduct in-depth research, so I anticipate that search will continue to serve as a supplement as more people use Bard."

Google only introduced Bard in February, unveiling the conversational AI in response to ChatGPT's meteoric rise in popularity. Microsoft went one step further, disclosing and launching Bing Chat, which is powered by OpenAI's GPT-4 language model, a few days after Google's announcement.

In contrast to these two systems, Bard is based on Google's own language model, LaMDA, which made headlines in June 2022 after an engineer, Blake Lemoine, was placed on administrative leave for publishing transcripts he claimed proved the system was intelligent.

The company has been a pioneer in AI technology for many years, even devising the "transformer" technology that became the T in "GPT" in 2017. However, the company has historically struggled to deliver products based on this research, which insiders attribute to a combination of organizational dysfunction and a concern that AI technology could threaten the company's profitable core businesses.

Google did not respond to an inquiry about how much more it costs to process a Bard query than a conventional search, instead focusing on "efficiency improvements" the company had made. Cost estimates from outside sources range from a 10- to 100-fold increase, yet Bard has no advertisements and no apparent revenue sources.

Although Google is playing catch-up to would-be AI disruptors, these companies have their own obstacles to overcome before they can produce software suitable for millions of users.

On Monday, a privacy violation at OpenAI allowed users to view the chat histories of others. The service was taken offline while the company dealt with the security compromise, and as of Tuesday morning, many users were still unable to access the history function.

ALSO READ: Google Bard AI Login: How to Use, Launch Date, Bard Ai Signup

Such defects could hinder the company's efforts to position itself as a viable service provider for corporate clients like PwC, which last week signed an international agreement to use Harvey, an OpenAI-backed legal chatbot. OpenAI did not respond to a request for comment regarding the security breach.