EWC urges the AI Office to not forget the sources of all AI development in the AI Pact: Authors, Artists and Performers

The AI Act Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence entered into force on August 1, 2024. In this context, the European Commission is promoting the “AI Pact”, seeking the GPAI and AI industry’s voluntary commitment and compliance to follow the AI Act and to start implementing its rules ahead of the legal deadline(s). EWC took part in the consultation of selecting topics in AI Pact seminars and made proposals for further activities and involce especially authors, artists and performers within the AI Pact exchanges.

On July 22, 2024, the European Commission published the draft AI Pact pledges pursuant to the EU Artificial Intelligence Act (the EU AI Act) and opened a consultation until 18 of August 2024 on the topics to be selected for further seminars related to the drafting of the AI Pact.

The EWC contributed within the consultation, quote:

From the perspective of one of the core groups negatively affected by GPAI and generative informatics, we, in representation of 50 organisations of writers and translators in the book sector, representing 220,000 individual authors of book works in 32 EU and non-Eu countries, kindly ask to acknowledge the following:

Currently, the aims of the AI pact are focused on the workflow and interests of AI developers, deployers, and providers, incl. on technical means. The implementation of the AI Act needs the holistic view on the full AI ecosystem:

(a) The sources of any so called „data“ – individuals, but in forefront writers, authors, artists and performers; (b) the collectors of copyright-protected works – corpora builders, sets-curators; and (c) the consumers and passive users of (G)AI products or AI-based decisions: the public.

In addition, it is necessary to develop a functioning technical system that makes both input and output traceable. It is also necessary to communicate to AI developers that machine-readable watermarking can be easily removed, and that GAI outputs require human-readable labelling. This is essential to enable fully informed decision-making and reception, as well as to avoid future (and occurring) remuneration fraud.

It is essential that the AI Pact helps to establish an end-to-end chain of responsibility and liability in which all parties involved are always informed and act in accordance with the law.

Finally, it is of great importance that AI developers realise that the design of the CDSM Directive Art 4 (as well as 3 and its implications of private partnerships between research institutions and commercial GPAI developers) is so controversial that a business model can by no means be built on the one-sided interpretation that Text and Data Mining is the same as GPAI and GenAI development, alone. Licensing and the respect for the decisions of all those who originate works are just as much a part of AI literacy as understanding of intellectual property law.

To the full EWC contribution:  AI Pact EWC Contribution 240816

Background of the AI Pact

The European AI Office, established in February 2024,  has initiated the development of the AI Pact, to share ideas through workshops organised by the EU executive’s AI Office among the so far 700 already registered AI businesses. The AI Pact is structured around two pillars, quote from the website https://digital-strategy.ec.europa.eu/en/policies/ai-pact:

• Pillar I acts as a gateway to engage the AI Pact network (those organisations that have expressed an interest in the Pact), encourages the exchange of best practices, and provides with practical information on the AI Act implementation process;
• Pillar II encourages AI system providers and deployers to prepare early and take actions towards compliance with requirements and obligations set out in the legislation.

The AI Pact shall allow AI businesses to plan ahead and prepare for the full implementation of the AI Act. This is a part of the European AI Alliance, a European Commission initiative, to establish an open policy dialogue on AI.

Background: The European AI Office

Description from https://digital-strategy.ec.europa.eu/en/policies/ai-office

“At an institutional level, the AI Office works closely with the European Artificial Intelligence Board formed by Member State representatives and the European Centre for Algorithmic Transparency (ECAT) of the Commission.

The Scientific Panel of independent experts ensures a strong link with the scientific community. Further technical expertise is gathered in an Advisory Forum, representing a balanced selection of stakeholders, including industry, startups and SMEs, academia, think tanks and civil society. The AI Office may also partner up with individual experts and organisations. It will also create fora for cooperation of providers of AI models and systems, including general-purpose AI, and similarly for the open-source community, to share best practices and contribute to the development of codes of conduct and codes of practice.

The AI Office also oversees the AI Pact, which allows businesses to engage with the Commission and other stakeholders such as sharing best practices and joining activities. This engagement has started before the AI Act entered into force and will allow businesses to plan ahead and prepare for the full implementation of the AI Act. All this will be part of the European AI Alliance, a Commission initiative, to establish an open policy dialogue on AI.

Further initiatives to foster trustworthy AI development and uptake within the EU are mapped on the Coordinated Plan on AI.”