IFLA took part in the second session of the conversation on Intellectual Property and Artificial Intelligence organised by the World Intellectual Property Organization (WIPO) on 7-9 July 2020.

After the first session on intellectual property and artificial intelligence in September 2019 and due to the COVID-19 situation, WIPO organised the conference in virtual form, in order to discuss issues related to this topic further. Many contributors from all around the world took this opportunity to share their views.

Organised over three days, the conversation centred on topics highlighted in the revised issues paper on intellectual property policy and artificial intelligence, published in May 2020.

The agenda included themes such as IP protection for AI-generated and AI-assisted works and inventions and related topics developed and comments on the notion of inventorship and ownership, the notion of authorship and ownership, deep fakes, copyright policy issues and copyright in training data, further rights in data and trade secrets.

IFLA underlined that legislators should avoid premature regulation of artificial intelligence, given that this area is still evolving rapidly. Any rules passed might become ineffective or harmful to further developments.

Regarding the right to use legally accessed materials for text-and-data mining and training machine learning, IFLA highlighted the necessity for an enforceable cross-border exception, in order to support innovation and provide legal certainty to libraries.

IFLA also raised, once again, one of its primary concerns regarding user rights.

Recent legislation has shown overconfidence in digital tools as a means of identifying when a use of a work counts as copyright infringement, and when an exception of limitation is in play. A balanced regulation system cannot rely on technology alone..

Recordings of the meetings are available:

Day 1: IP Protection for AI-generated and AI-assisted works and inventions and related topics.

Day 2: AI inventions: Patentability, disclosure and guidelines

Day 3: Data: Copyright in Training Data, Further Rights in Data and Trade

You can also read IFLA’s statements:

Session 1: IP protection for AI-generated and AI-assisted works and inventions and related topics

Session 3: Data: Copyright in Training Data, Further Rights in Data and Trade Secrets