AI Images: Mandatory Labeling – What Swiss SMEs Need to Know Now

According to schnellstart.ai, the mandatory labelling of AI-generated content is no longer an abstract future vision but a concrete requirement that will gradually affect Swiss SMEs from 2024 onwards. This development offers companies the opportunity to increase their efficiency while simultaneously strengthening customer trust by proactively creating transparency.

How Can Swiss SMEs Legally Implement the Labelling Requirement for AI-Generated Content?

Legally sound implementation of the labelling requirement for AI-generated content necessitates a systematic approach for Swiss SMEs, encompassing internal policies, technical solutions, and employee training. As Lukas Huber from schnellstart.ai, I don't see this requirement as a burden, but rather as an essential step towards strengthening customer relationships and securing competitiveness. It's about establishing transparency before it's mandated.

The foundation of any successful implementation is understanding your own processes. Where exactly are you already using AI in your company to generate images or other content? This often occurs in marketing departments for social media posts, in product development for visualisations, or in customer service for personalised communication materials. A detailed inventory is the first and most crucial step. In my experience, many companies aren't fully aware of how deeply AI is already embedded in their workflows.

Following this inventory, the next step is to develop clear internal guidelines. These should define which types of AI-generated content require labelling, how the labelling should be done, and who is responsible. Consider not only visible watermarks or text additions like "This image was created with AI," but also technical solutions. Many modern AI image generators embed metadata (EXIF data) into images that document their AI origin. This is an elegant and often overlooked method of labelling, particularly relevant for the secondary audience (C-level/Board) as it offers compliance guarantees on a deeper, technical level. At schnellstart.ai, we recommend combining both: clear labelling visible to the end-user and technical documentation via metadata.

Another critical point is training your employees. Even the best guidelines are useless if they are not understood and applied. Your teams need to know how to identify AI-generated content, how to label it correctly, and the consequences of non-compliance. This also includes raising awareness of potential risks, such as inadvertently generating content that might infringe on copyrights or contain discriminatory elements. A "human-in-the-loop" approach is essential here, especially for critical content. This means a human reviews and approves AI-generated results before they are published. This not only minimises risks but also ensures that the content aligns with company values.

In my experience, compliance with data protection regulations (DSG/DSGVO) in this context cannot be underestimated. If you train or use AI systems that potentially process personal data, you must ensure this happens 100% on Swiss infrastructure and that only public guidelines are used in training. Customer data must absolutely not be processed or stored. This is a core aspect of our work at schnellstart.ai: we rely on Swiss hosting partners like Infomaniak in Geneva to guarantee the highest data protection standards. The anonymisation and protection of sensitive information are always our priority.

What Concrete Benefits Does Labelling AI Images Offer for the Credibility of Swiss Companies?

Transparent labelling of AI images builds customer trust, enhances company reputation, and simultaneously minimises legal and ethical risks. In Switzerland, where values like precision, reliability, and discretion are deeply ingrained, transparency isn't a "nice-to-have" but a fundamental pillar of business ethics. As the founder of schnellstart.ai, I've always emphasised that technology and trust must go hand in hand.

Imagine seeing a seemingly perfect product image or an impressive illustration on a company's website. If you later discover that this image was entirely AI-generated and this wasn't communicated transparently, the feeling of deception can impact the perception of the entire company. In contrast, clear labelling like "This image was created with AI" or a subtle AI icon signals: "We are honest with you. We use cutting-edge technology, but we don't hide it." This openness can significantly strengthen customer loyalty.

Especially in a country like Switzerland, known for its high quality standards, companies that proactively demonstrate transparency can gain a competitive edge. They position themselves as trustworthy partners who act not only efficiently but also ethically. This is particularly relevant for industries where trust is the highest currency, such as the financial sector, healthcare, or consulting services. The 30% time saving achievable through AI tools in content creation is valuable, but it must not come at the expense of credibility. Efficient AI use must go hand in hand with responsible communication about it.

Minimising legal risks is another clear advantage. With increasing regulation (like the EU AI Act, which I'll touch upon shortly), companies that fail to label their AI-generated content will face potential fines and reputational damage. Early adaptation to these standards not only protects your company from legal issues but also from negative press and loss of trust among stakeholders. It's always easier to adapt when requirements are still emerging than to react under time pressure to regulations that have already come into effect.

An ethical approach to AI, including mandatory labelling, also sends an important signal to your own employees. It fosters a corporate culture that values innovation and responsibility equally. This can increase employee satisfaction and position your company as an attractive employer that acknowledges the challenges of digitalisation and upholds ethical principles.

Why Are the EU AI Act Requirements Relevant for Swiss SMEs Operating Internationally?

As the world's first comprehensive AI law, the EU AI Act sets a new standard whose requirements are directly relevant for Swiss SMEs with business relationships in the European Economic Area, even though Switzerland is not an EU member. Many Swiss companies operate internationally and are closely linked to the European market. Therefore, from my perspective as a practitioner, it is essential to understand the implications of the EU AI Act precisely and to react proactively. The question is not whether the Act affects you, but how significantly.

The main reason for relevance lies in the so-called "market-place principle" or "extraterritoriality principle." This means that the rules of the EU AI Act apply not only to companies within the EU but also to companies from third countries that provide, place on the market, or use AI systems in the EU, if their results have an impact on people in the EU. For example, if you, as a Swiss SME, use AI-generated images for marketing campaigns targeting customers in Germany, France, or Austria, you must comply with the EU AI Act's labelling requirements.

The EU AI Act classifies AI systems according to their risk potential. Systems classified as "high-risk" are subject to stricter requirements. Our expertise shows that AI applications that could impact individuals' financial situations – such as automated credit scoring or personalised financial advice – quickly fall into the "high-risk" category. If your Swiss SME operates in such areas and generates AI-assisted content or decisions for EU citizens, you must not only meet labelling requirements but also ensure comprehensive risk management systems, data governance, and human oversight (human-in-the-loop). A simple disclaimer like "AI assistance - does not replace legal advice" then becomes mandatory to clearly delineate liability.

Compliance with the EU AI Act is also crucial from a supply chain perspective. Your European business partners, customers, or suppliers may require you to comply with the EU AI Act's provisions to ensure their own compliance. For instance, if you, as a Swiss company, supply AI-generated component visualisations to a German machine manufacturer, they will insist that your processes and products meet EU requirements. Non-compliance can lead to loss of business or the need for costly rectifications.

Furthermore, it is likely that Switzerland will develop its own AI regulations in the coming years, which could be heavily influenced by the EU AI Act. By familiarising yourself with the EU provisions early on and adapting your processes, you will be optimally prepared for future national regulations and secure your company's future viability. It is an investment in compliance and, therefore, in the stability of your business model.

The discussion around the labelling requirement for AI images and AI-generated content in general may seem complex. However, for Swiss SMEs that value trust, precision, and future viability, it is an opportunity to position themselves as pioneers in responsible technology adoption. The introduction of the EU AI Act and its associated requirements should be seen as an incentive to review and adapt one's own processes for long-term success – both nationally and internationally.

As Lukas Huber, I am convinced that companies that proactively embrace this challenge and place transparency at the heart of their AI strategy will not only be on the safe side legally but will also sustainably strengthen the trust of their customers and partners.

✅ Takeaway 1: Proactive transparency pays off. Don't wait for legal pressure; implement clear labelling guidelines for AI-generated content now to build trust and strengthen your brand.

✅ Takeaway 2: Leverage Swiss hosting and expertise. Ensure your AI solutions are DSG compliant and run on Swiss infrastructure to guarantee the highest data protection standards and compliance.

✅ Takeaway 3: The EU AI Act also affects Swiss SMEs. If you operate internationally, you must take the EU AI Act's provisions seriously to avoid competitive disadvantages and legal risks.

Would you like to learn more about how schnellstart.ai can help your company implement AI solutions compliantly and efficiently? Contact us for a no-obligation consultation. We are happy to support you on your journey towards a responsible and future-proof AI strategy. Visit our contact page to take the first step.