Artificial intelligence for creative workers
A TUC manifesto
Introduction
This manifesto outlines our values and proposals for addressing the impact of artificial intelligence (AI) on creative work and workers. While we focus on the creative and education sectors, our manifesto advocates for the interests of all workers who generate intellectual property or use their likeness while at work, including writers, performers, educators and others.
The creative and education sectors are vital to the UK’s economy, social cohesion and cultural identity. Creative workers are key contributors to AI development but face economic vulnerability if their rights are not protected. The rapid development of AI technology requires a response which centres the rights and interests of creative workers.
Values
These are the values we consider fundamental to ensure this technology benefits all:
- Transparency: Technology companies should provide clear information about how the technology operates and the data it is trained on. This is crucial for informed consent and preventing misinformation.
- Consent and agency: People should be able to decide on how they engage with this technology and should be able to withdraw consent.
- Human creativity and connection: Human creativity has inherent value and should be safeguarded. Human input is essential for the quality, authenticity and emotional resonance of creative work.
- Rights protection and preservation: The development and deployment of technology should respect, preserve and support workers’ rights and intellectual property rights.
- Benefit-sharing, compensation and remuneration: The gains from this new technology should be shared fairly with workers so they are compensated and can continue to contribute.
- Technology ‘for-and-by’ creative workers: AI should be designed with creative workers to meet their needs.
- Training and skills development: Digital literacy and traditional skills training are essential for safe and effective technology use.
- Consultation and collective representation: Creative workers and unions should be involved in technology design and deployment decisions.
- Collaboration: Increased collaboration between technology stakeholders and creative workers is needed.
- Equality, inclusion and cultural diversity: Technology should uphold equality, inclusion, and cultural diversity, avoiding content homogenisation.
Proposals
These proposals take forward the values outlined above:
- Labelling of machine-generated outputs: Clear labelling should be in place to differentiate machine-generated outputs from human-created content.
- Opt-in for data mining and AI training: The use of human-generated materials should only be permitted if creative workers and rights holders have given their permission and consent.
- Right to remove content from training datasets: The government should recognise a right to remove content from training datasets and ensure there are clear enforcement routes.
- Fair contracts: Safeguards should be put in place against unfair terms and practices in contracts, which often arise because creative workers can be vulnerable in contractual negotiations.
- Preserved and increased intellectual property rights: Government should confirm and uphold the principle that data mining for AI training without consent is an infringement of intellectual property rights and increase protection for creative workers in response to the new technology.
- New likeness rights: New rights should be implemented to protect workers’ likenesses from being used without their consent, such as in ‘deepfakes’.
- Remuneration schemes and licensing agreement: There should be licensing and compensation mechanisms for data mining and AI training so that they can be carried out legally with informed consent and fair remuneration of workers.
- Credits and rights communication: Workers should be clearly and consistently attributed for their work when it is used by technology companies.
- Disclosures: Technology companies should clearly disclose how their technology operates and what data it has been trained on.
- Accessible legal redress: There should be user-friendly and timely fora for rights enforcement, backed up with strong sanctions such as fines.
- Harmonised protection of creative workers: Governments should collaborate to prevent regulatory disparities which could encourage ‘jurisdiction shopping’.
- Independent AI regulatory body: The government should establish a regulatory body, with social partner representation, to oversee and regulate the deployment of AI.
- Support for sector-specific and rights-compliant AI: There should be co-operation between workers’ unions, technology leaders and government to support AI technology tailored to the usage needs of creative workers.
- Specialised training and guidance: The government, further and higher education organisations, as well as unions should provide training opportunities for creative workers on data, technology and rights relevant to new technologies like AI. This training should be sector-specific.
Stay Updated
Want to hear about our latest news and blogs?
Sign up now to get it straight to your inbox