Don’t let tech giants steal copyrighted content to train their artificial intelligence chatbots, Lords says

Artificial intelligence companies should not continue to steal copyrighted material to train their chatbots, a House of Lords report has warned.

Colleagues highlighted their “deep concerns” about tech companies sucking up content from books and news websites on an “absolutely enormous scale”.

The House of Lords communications and digital committee said ministers have a ‘duty’ to prevent tech giants from taking control of the multi-billion pound AI industry, warning that urgent safeguards were needed.

The rise of ChatGPT has increased demand for the technology, with millions of people now using the tools every day, from writing school essays to drafting legal opinions.

The rise of ChatGPT has increased demand for the technology, with millions of people now using the tools every day, from writing school essays to drafting legal opinions (File image)

News publishers warned that AI tools could make it impossible to produce independent journalism.

Open AI – the company behind ChatGPT – Meta and Microsoft all suggested it was virtually impossible not to infringe on copyright, claiming that restricting access would lead to worse AI models.

The report says the government ‘cannot sit on its hands for the next decade and hope that the courts will provide an answer’.