For more than three months, Google executives have watched as projects by Microsoft and a San Francisco startup called OpenAI have fueled the public’s imagination with the potential of artificial intelligence.
But on Tuesday, Google tentatively stayed on the sidelines when it launched a chatbot called Bard. The new AI chatbot will be available to a limited number of users in the United States and Britain and will adapt to additional users, countries and languages over time, Google executives said in an interview.
The cautious launch is the company’s first public effort to address the recent chatbot craze fueled by OpenAI and Microsoft, and is meant to show that Google is capable of providing similar technology. But Google is taking a much more cautious approach than its competitors, which have faced criticism that unpredictable and sometimes unreliable technology is proliferating.
Still, the launch represents a significant step toward averting a threat to Google’s most lucrative business, its search engine. Many in the tech industry believe that Google, more than any other big tech company, has a lot to lose and a lot to gain from AI, which could help make a range of Google products more useful, but it could also help other companies to participate in Google’s huge business. internet search business. A chatbot can instantly produce complete sentence responses that don’t force people to scroll through a list of results, which is what a search engine would.
Google started Bard as a web page itself rather than a component of its search engine, beginning a tricky dance of embracing new AI while preserving one of the tech industry’s most profitable businesses.
“It’s important that Google start playing in this space because that’s where the world is headed,” said Adrian Aoun, a former director of special projects at Google. But the move to chatbots could help change a business model that relies on advertising, said Aoun, who is now the chief executive of healthcare startup Forward.
Inside the world of big technology
In late November, San Francisco start-up OpenAI launched ChatGPT, an online chatbot that can answer questions, write term papers, and talk about almost anything. Two months later, the company’s main investor and partner, Microsoft, added a similar chatbot to its Bing Internet search engine, showing how the technology could change the market that Google has dominated for more than 20 years.
Google has been racing to ship artificial intelligence products since December. Declared a “code red” in response to the release of ChatGPT, making AI a core priority for the company. And it prompted teams within the company, including researchers who specialize in studying AI security, to collaborate to speed approval of a wave of new products.
Industry insiders have wondered how quickly Google can develop a new AI technology, particularly given the breakneck pace of OpenAI and Microsoft releasing their tools.
“We are in a singular moment,” said Chirag Dekate, an analyst at technology research firm Gartner. ChatGPT inspired new start-ups, captured the public imagination and sparked more competition between Google and Microsoft, he said, adding: “Now that market demand has changed, Google’s approach has too.”
Last week, OpenAI tried to up the ante with a newer technology called GPT-4, which will allow other companies to build the kind of artificial intelligence that powers ChatGPT into a variety of products, including trading software and trading websites. electronic.
Google has been testing the technology behind Bard since 2015, but hasn’t released it beyond a small group of early testers so far because, like the chatbots offered by OpenAI and Microsoft, it doesn’t always generate reliable information and can show bias. against women and women. people of color
“We are very aware of the problems; we need to responsibly bring this to market,” said Eli Collins, Google’s vice president of research. “At the same time, we see all the enthusiasm in the industry and the enthusiasm of all the people using generative AI”
Mr. Collins and Sissie Hsiao, Google’s vice president of products, said in an interview that the company had not yet determined a way to make money from Bard.
Google announced last week that AI would be coming to its productivity apps like Docs and Sheets, which businesses pay to use. The underlying technology will also be for sale to companies and software developers who want to create their own chatbots or power new applications.
“It’s early days for technology,” Ms. Hsiao said. “We are exploring how these experiences can appear in different products.”
The recent announcements are the start of Google’s plan to introduce more than 20 artificial intelligence products and features, The New York Times reported, including a feature called Shopping Try-on and the ability to create custom background images for YouTube videos and Pixel phones.
Instead of being combined with your search engine, Bard is a stand-alone web page that presents a question box. At the bottom of a reply is a “Google it” button, which takes users to a new tab with a conventional Google search results page on the topic.
Google executives pitched Bard as a creative tool designed to compose emails and poems and offer guidance on how to engage kids in new hobbies like fly-fishing. The company is eager to see how people use the technology and will further refine the chatbot based on usage and feedback, the executives said. Unlike your search engine, however, Bard was not primarily designed to be a reliable source of information.
“We thought of Bard as an add-on to Google Search,” said Ms. Hsiao. “We want to be bold in the way we innovate with this technology, as well as be responsible.”
Like similar chatbots, Bard is based on what’s called a large language model, or LLM, a type of artificial intelligence technology that learns skills by analyzing vast amounts of data from the Internet. This means that the chatbot is often wrong and sometimes makes up information without warning, a phenomenon AI researchers call hallucination. The company said it had worked to limit this behavior, but acknowledged that its controls were not entirely effective.
When executives demoed the chatbot on Monday, it refused to answer a medical question because doing so would require accurate and correct information. But the bot also falsely described the source of a response it generated about the American Revolution.
Google posts a disclaimer in the Bard query box warning users that problems may arise: “Bard may display inaccurate or offensive information that does not represent the views of Google.” The company also gives users three answer options for each question and allows users to provide feedback on the usefulness of a particular answer.
Like Microsoft’s Bing chatbot and similar bots from startups like You.com and Perplexity, the chatbot notes its responses from time to time, so people can review their sources. And it matches Google’s index of all websites, so you can get instant access to the latest information published on the Internet.
This can make the chatbot more accurate in some cases, but not all. Even with access to the latest information online, it still misrepresents the facts and creates misinformation.
“LLMs are complicated,” said Mr. Collins, Google’s vice president of research, referring to the technology that underpins chatbots today. “Bard is no exception.”
