By Si Barber

In February 2025 the UK government initiated discussions around potential changes to copyright law to accommodate the rise of artificial intelligence (AI), particularly focusing on how AI systems use copyrighted material to learn from and train its responses.

While no final legislation has been enacted yet, these proposals stem from a consultation process launched earlier in the year, with input from various stakeholders, including the creative industries. 

The link to the official consultation is below and you are encouraged to make your thoughts known before the deadline of 25^th February.

https://ipoconsultations.citizenspace.com/ipo/consultation-on-copyright-and-ai/consultation/

Below is an overview of the proposed changes and their potential impact on photographers, based on the current discourse and developments.

Proposed Changes to UK Copyright Law

The primary focus of the proposed changes involves introducing a copyright exception for AI training. This would allow AI developers to use copyrighted works,including photographs, without needing explicit permission from the copyright holders, provided the use is for training AI models. Key elements include:

1. Exception to Copyright for AI Training: Unlike the current framework, where permission or licensing is typically required to use copyrighted material, this exception would permit AI companies to scrape and process images (and other works) from the internet or other sources for AI development, without compensating or notifying the original creators.

2. Opt-Out Mechanism: One suggested approach is an opt-out system, where creators would need to explicitly indicate that their works cannot be used for AI training. This shifts the burden from AI companies (to seek permission) to creators (to protect their work).

3. No Mandatory Registration, But Potential Costs: While copyright in the UK remains automatic upon creation (no formal registration is required), an opt-out system might encourage or necessitate photographers to take additional steps—such as watermarking, metadata tagging, or registering with a database—to enforce their preferences, potentially incurring costs or administrative effort.

4. Transparency and Redress: Following pushback from creative industries, there have been calls for greater transparency from AI firms about the datasets they use and mechanisms for creators to seek redress if their work is exploited. Amendments proposed in the House of Lords in January 2025 aim to strengthen these protections, though their final inclusion is uncertain.
   
   
   

How These Changes Could Affect Photographers

Photographers, whose work is inherently visual and widely shared online, stand to be significantly impacted by these changes. Here’s how:

1. Loss of Control Over Work:

   • Currently, photographers automatically own the copyright to their images upon creation, giving them exclusive rights to control reproduction, distribution, and public display. An AI training exception could allow companies to use these images without permission, undermining this control.

   • For example, wedding photos, stock photography, or artistic portfolios posted online could be ingested into AI systems to generate derivative works, potentially without the photographer’s knowledge or consent.

2. Economic Impact:

   • Photographers often rely on licensing fees for income, especially in commercial or stock photography. If AI firms can use their images for free under a copyright exception, this could reduce demand for licensed content, as AI-generated alternatives flood the market.

   • AI companies are generally based in the USA and China, as such the loss of income will be on British Tax payers. A change in legislation risks taking money out of the UK economy and giving it to foreign tech companies, a continuation of a long standing trend which will diminish the tax-take to the Exchequer.

   • High-profile creators like Sir Paul McCartney have voiced concerns about a “Wild West” scenario where artists lose economic benefits to tech companies profiting from their work.

3. Administrative Burden:

   • An opt-out system would require photographers to proactively mark their images as excluded from AI training—potentially through metadata, watermarks, or third-party registries. For professionals managing tens of thousands of images (e.g., event or wildlife photographers), this could become time-consuming and costly, especially if fees are introduced for opt-out mechanisms or verification services.

4. Moral Rights and Attribution:

   • UK law grants photographers moral rights, including the right to be identified as the author and to object to derogatory treatment of their work. AI-generated outputs might strip away attribution or alter images in ways photographers find objectionable, yet enforcing these rights could become harder if the initial use (training) is legally permitted.

5. Competition from AI-Generated Content:

   • AI trained on photographers’ images could produce similar works, competing directly with human creators. This might devalue original photography, particularly in markets like advertising or editorial content, where AI alternatives could be cheaper and faster to produce.

6. Legal Recourse:

   • If an exception is enacted, photographers might find it harder to pursue infringement claims against AI companies, as the use would be legally sanctioned. However, proposed amendments for transparency and redress could offer some avenues to challenge misuse, though these would likely require evidence of specific harm.

Current Status and Industry Response

The consultation period for these proposals closes on February 25, 2025, meaning the details are still under negotiation. The creative sector, including groups like the Creative Rights in AI Coalition (CRAIC)—which represents photographers, musicians, and other artists—has strongly opposed a broad copyright exception. A House of Lords vote on January 28, 2025, supported amendments to the Data (Use and Access) Bill to bolster creator protections, signalling significant resistance to weakening copyright law. Industry leaders argue that AI should enhance, not exploit, human creativity, advocating for a licensing model over an exception.

Potential Outcomes for Photographers

• Best-Case Scenario: If creator-friendly amendments are adopted, photographers might retain more control, with AI firms required to license content or provide clear opt-in/opt-out mechanisms at no cost to creators.

• Worst-Case Scenario: A broad exception with a weak opt-out system could flood the market with AI-generated imagery derived from photographers’ work, reducing their income and influence over their creations.

In summary, while the proposed changes aim to balance AI innovation with creative rights, they currently risk shifting power away from photographers toward tech companies. The final impact will depend on how the government responds to the consultation and whether it prioritizes the economic and moral rights of creators like photographers. For now, photographers may want to monitor developments closely, consider embedding protective metadata in their images, and engage with advocacy groups to influence the outcome.

©Si Barber. Moral rights asserted.