Google Tests A.I. Tool That Is Able to Write News Articles

Google is testing a product that uses artificial intelligence technology to create news stories and is pitching it to news outlets including The New York Times, The Washington Post and News Corp, which owns The Wall Street Journal, according to three people familiar with the matter.

The tool, known internally under the working title of “Genesis,” can capture information such as details of current events and generate news content, said the people, who spoke on condition of anonymity when discussing the product.

One of three people familiar with the product said Google believes it can act as a kind of personal assistant for journalists, automating some tasks and saving others time, and sees it as a responsible technology that can help keep the publishing industry away from the pitfalls of generative AI.

Some executives who saw Google’s pitch, asking for anonymity to discuss confidential matters, said it was fear-mongering. The two said that the effort put into producing accurate and artistic news stories seemed to be taken for granted.

“We are in the early stages of exploring the idea of ​​potentially partnering with news publishers, particularly smaller publishers, to provide AI-enabled tools to assist journalists in their work,” Google spokeswoman Jen Crider said in a statement.

“Simply put, these tools are not intended to, and cannot, replace the crucial role that journalists play in reporting, producing and fact-checking stories,” she added. Alternatively, you can provide headings and other writing options.

“We have a good relationship with Google and appreciate Sundar Pichai’s long-term commitment to journalism,” a News Corp spokesperson said in a statement.

The Times and Post declined to comment.

Jeff Jarvis, a professor of journalism and media commentator, said there are potential upsides and downsides to Google’s new tools like the one above.

“Journalists should use this tool if it can reliably deliver factual information,” said Jarvis, director of the Tonite Center for Entrepreneurial Journalism at the Craig Newmark School of Journalism at the City University of New York.

“On the other hand, abuse of this tool by journalists and news outlets on subjects that require nuance and cultural understanding could undermine not only the tool but also the credibility of the news outlets that use it,” he continued.

News outlets around the world are grappling with whether to use artificial intelligence tools in their newsrooms. A number of companies, including The Times, NPR and Insider, have informed employees that they will explore potential uses for AI as they consider how it can be applied responsibly to the high-stakes news arena where seconds and accuracy are paramount.

But Google’s new tool is sure to stir fears even among journalists who’ve been writing their own stories for decades. Some news outlets, such as the Associated Press, have long used AI to generate stories about things like corporate earnings reports, but the service’s coverage remains small compared to articles written by journalists.

Artificial intelligence could change this situation, allowing users to create a wider range of articles, which, if not carefully edited and checked, can spread misinformation and affect perceptions of traditionally written articles.

While Google is working at breakneck speed to develop and deploy generative AI, the technology also poses some challenges for the advertising giant. Google has traditionally been responsible for organizing information and sending users to publishers’ websites to read more, but tools like its chatbot Bard present factual claims that are sometimes inaccurate and don’t direct traffic to more credible sources such as news publishers.

The technology comes as governments around the world ask Google to give more of its advertising revenue to news outlets. After the Australian government tried to force Google to negotiate with publishers over payments in 2021, the company further stepped up its partnerships with international news outlets under its News Showcase program.

Publishers and other content creators have already criticized Google and other big AI companies for using decades of articles and posts to train AI systems without compensating publishers. News outlets such as NBC News and The Times have taken a stand against AI siphoning data without permission.