Hong Kong copyright law changes in pipeline to keep pace with artificial intelligence development
A review of Hong Kong's outdated copyright regulations is under way to keep pace with the rapid development of artificial intelligence (AI) and experts in the field have backed changes to legislation to protect creators who use the technology.
David Wong Fuk-loi, the director of the government's Intellectual Property Department, mapped out a consultation plan with online service providers on topics involving copyright of AI-generated content, machine-learning answers and models, and the protection of AI content creators, which are not covered by existing laws.
"In response to the prevalence of AI, we are to release a consultation paper within this year and review what other areas in the current ordinance need to be amended," Wong said last week.
Do you have questions about the biggest topics and trends from around the world? Get the answers with SCMP Knowledge, our new platform of curated content with explainers, FAQs, analyses and infographics brought to you by our award-winning team.
Wong added that AI development had built up momentum around the world, but the city's copyright regulations had not kept up.
He said the Copyright Ordinance could protect original material created by a human, but AI-generated content had complicated the definition of ownership.
Wong added that the process of training AI to source content from the internet also constituted another copyright problem, as the technology could learn to produce or replicate information and patterns created by a human.
"These legal problems have been widely discussed overseas and we have been closely monitoring them," he said.
Wong added that this year was the right time for the department to draft a consultation paper.
But he said the department had not yet made any firm suggestions on what could be added to the Copyright Ordinance to modernise it.
A European Parliament vote on proposed AI legislation is scheduled for April after two committees earlier endorsed a provisional law to ensure that the level of AI use by government and private enterprises was in line with the protection of fundamental rights.
The European bill was designed to regulate AI based on its capacity to cause harm to society under a "risk-based" approach, or the higher the risk, the stricter the rules.
Several tech giants and software developers - including OpenAI, GitHub, Meta and Google - have faced legal action related to copyright infringement.
The New York Times has launched legal action against OpenAI and Microsoft over what it said was the unauthorised use of its published work to help create artificial intelligence systems.
The newspaper argued that millions of its articles were used to train automated chatbots designed to compete with it as an information source.
Francis Fong Po-kiu, the honorary president of the Hong Kong Information Technology Federation, said a lot of content on internet platforms was not regulated or covered by the Copyright Ordinance.
He added subjects such as whether a bot had the right to generate content from material under copyright were grey areas in the city's legislation.
"AI has been widely applied in recent years, but [the law in] Hong Kong does not get with the times," Fong said.
"The speed of AI development is unpredictable. The law we are now discussing to tackle those relevant legal issues is only applicable for the time being."
Fong added that creation of a new law was a long process and it was a clear case of "the sooner the better" for public consultation on the subject.
Alan Chiu Chi-wai, a lawyer and specialist in intellectual property law, agreed that there was a need to plug gaps in copyright legislation.
He explained that the legislative intent for copyright was to protect creativity, but at the moment it was impossible to clarify whether work was from human intellectual input or AI.
He added that legal problems generated by the uncertainty had confused users on what crossed the line into plagiarism, even if it was done unintentionally.
"Most importantly, the law should clarify whether copyright subsisted in AI-generated things," Chiu said. "Once that is defined, the next question would be 'who owns it?'."
He added many of the people involved - such as AI users, digital creators and software developers - had a variety views on AI copyright, so he did not think there was a quick fix.
"I think [authorities] need to decide whether they want to go into the details of how an amendment would turn out to be, or put forward a legal framework to let the industry know that there are legal protections to develop AI business in Hong Kong," Chiu said.
Sandra Hui Lai-shan, the assistant director of intellectual property (copyright), on Friday said authorities also planned to replace the present permitted amount of copying of published works in public libraries with up to 10 per cent of the piece.
The government on Thursday launched a two-month public consultation on the proposed subsidiary legislation under the Copyright Ordinance, designed to cover libraries, museums and archives.
The proposed changes outline conditions for certain permitted acts in the use of copyrighted works.
People can at present copy a maximum of 4,000 words in an article, and a total of 8,000 words in a series of extracts with each at most 3,000 words under a "reasonable proportion" rule.
Hui explained that the government hoped to scrap the word-count approach to provide flexibility and would consider instances where the limit had been exceeded on a case-by-case basis.
"For example, in books, if we only allow copying an exact of 10 per cent of it, a sentence may be left incomplete," she told a radio programme. "[Residents] may need to copy a bit more, thus it really depends on the actual situation and the characteristics of the work."
Additional reporting by Edith Lin
This article originally appeared on the South China Morning Post (SCMP).
Copyright (c) 2024. South China Morning Post Publishers Ltd. All rights reserved.
