A Google artificial intelligence writing experiment conducted four years ago has resurfaced at the centre of a heated debate within the international literary community, highlighting how dramatically attitudes towards generative AI and copyright have shifted since 2022.
Thirteen professional writers who participated in Google’s experimental Wordcraft programme are now facing criticism from some authors and readers, with calls emerging online to boycott their books over their involvement in the project.
Google’s Wordcraft experiment returns to the spotlight
In 2022, Google invited 13 professional writers, including established science-fiction authors and New York Times bestselling writers, to participate in its Wordcraft Writers Workshop.
The project was run by Google’s DeepMind division and involved LaMDA, the language model that preceded technologies later used in Google Gemini.
Researchers wanted to understand whether artificial intelligence could assist professional writers with tasks such as brainstorming, gathering information and developing written material.
Participants received access to Wordcraft, a specialised text editor incorporating LaMDA. They were given eight weeks to produce a short story and could decide how extensively they wanted to use the AI system.
One participant chose not to include any AI-generated text. Google later published the results of the workshop alongside an academic research paper examining the experiment.
Attitudes towards generative AI changed rapidly
At the time, generative AI had not yet become the highly contentious issue it is across publishing and other creative industries.
That changed significantly following the rapid expansion of ChatGPT and competing generative AI platforms. Writers, artists, publishers and other rights holders increasingly raised concerns about copyrighted material potentially being used to train large language models without permission or compensation.
The debate has since developed into a major international copyright issue, accompanied by legal action against technology companies.
For New Zealand writers, publishers and other creative professionals, the dispute reflects wider questions being debated internationally about intellectual property protections as AI tools become increasingly capable of generating text, images and other creative material.
Old Google study triggers fresh criticism
The controversy surrounding Wordcraft intensified in early August after writer Lily Lachance discovered the 2022 study and shared it on Bluesky.
Lachance strongly criticised the 13 writers involved, arguing that their participation had helped Google legitimise technology she described as a “plagiarism machine”.
Her comments quickly attracted attention across the online literary community.
Some users subsequently called for readers to boycott books written by the participants and pushed for the writers involved in the experiment to be publicly identified.
The criticism was not focused solely on whether writers had used artificial intelligence. A central concern was the broader issue of how large language models obtain the enormous quantities of written material required for their training.
Participants defend decisions made in 2022
Several writers involved in the Wordcraft programme have responded to the renewed criticism by explaining the circumstances surrounding their participation.
Eugenia Triantafillou said she regarded Wordcraft in 2022 as an ordinary experiment involving an emerging technology and did not know the specific data used to train the underlying model.
She said her position on large language models has since changed and that she now holds a negative view of the technology. She would not participate in the same type of research today.
Other participants have similarly pointed to the substantial difference between public understanding of AI in 2022 and the debate surrounding the technology four years later.
Copyright concerns intensified after 2022
Major public discussion about generative AI, copyright and training datasets accelerated during 2023, following the widespread adoption of increasingly powerful AI systems.
That period also brought prominent legal challenges involving authors, publishers and technology companies, placing greater scrutiny on where AI developers obtained training material and whether copyright owners had provided permission.
As a result, actions that were previously viewed as experimentation with an emerging technology are now being reassessed through a substantially different ethical and legal lens.
Wordcraft controversy reflects wider creative industry divide
The renewed debate surrounding Google’s Wordcraft Writers Workshop illustrates the speed at which generative AI has transformed from an experimental technology into one of the most contentious issues facing creative industries.
For some writers, the 2022 workshop represents an early attempt to understand how AI might complement human creativity. For critics, however, participation helped advance technologies they believe could undermine writers’ livelihoods and intellectual property rights.
Four years after the experiment, the backlash demonstrates how quickly expectations surrounding AI, copyright and creative work have evolved — and why decisions made during the technology’s experimental phase continue to attract scrutiny today.
Alexander Donovan writes for News Collective, covering news, politics, business, technology, sport, entertainment, and lifestyle. He focuses on clear, reliable reporting and useful information, helping readers stay informed about current events, emerging trends, and stories that matter.