(San Francisco) Google is developing a new artificial intelligence (AI) tool to help journalists write their stories, in partnership with several big names in the press, the internet giant confirmed Thursday.
The New York Times had just revealed the information, stating that it was among the media involved, as were the Washington Post, the Wall Street Journal and other companies.
“We are in a phase of exploring ideas to potentially provide AI tools that will help journalists do their job,” a Google spokesperson said.
“These tools are not designed to — and cannot — replace the essential role journalists play in gathering and verifying information, and writing stories,” he said.
The idea is to design writing assistance features similar to those existing in Gmail or Google Docs, but specific for journalists, such as offering different titles or writing styles.
According to the New York Times, media executives have described Google’s project, dubbed “Genesis” internally, as “disturbing”.
OpenAI, the Californian startup that created ChatGPT and another leader in so-called “generative” AI, has also recently struck deals with news organizations.
The data-intensive company needed to train its language models has thus obtained permission to use the archives of the American news agency Associated Press (AP) since 1985.
In exchange, AP will have access to “OpenAI’s technology and expertise,” the two companies detailed in a statement last week.
OpenAI will also donate $5 million to the American Journalism Project, an organization that supports many local media outlets, and up to $5 million in credits to use its programming interface (API) to help journalists integrate AI tools into their production.
The phenomenal success of generative AI, capable of generating all kinds of text or images on a simple request in everyday language, is causing a lot of concern among content creators (artists, authors, journalists, etc.).
Several complaints have been filed against the companies concerned (including OpenAI).
Professionals accuse them of having used their content without consent or remuneration to feed their computer programs.
They also fear that this software will eventually become capable of replacing them.