Switching off AI on Fridays? What Swiss SMEs need to know about Copilot & Co.

The idea of completely switching off Artificial Intelligence on Fridays might sound absurd at first glance. An analyst proposed this provocative thesis to spark a debate about the uncontrolled use of AI tools like Microsoft Copilot. However, for Swiss SMEs, the question isn't whether to switch off AI, but how to integrate it intelligently and securely into their daily workflow to create genuine added value.

Especially in Switzerland, where SMEs form the backbone of the economy and often operate with limited resources, the strategic use of AI can lead to significant efficiency gains. A study shows that 45% of Swiss SMEs already view AI as an advantage for their business operations. However, blind faith in every new technology is just as dangerous as complete refusal. It's about control, governance, and a clear plan.

The challenge lies in unlocking potentials without introducing new risks – whether in data security, compliance, or employee productivity. This becomes particularly critical when implementation occurs without a clear strategy, and employees use the tools at their own discretion.

How can Swiss SMEs leverage the benefits of AI tools like Copilot without jeopardising productivity?

Through a clear strategy and sound implementation. Many SMEs see AI tools like Copilot as an opportunity to automate manual tasks and increase efficiency. This is fundamentally correct. Microsoft itself is experiencing increased demand for paid Microsoft 365 Copilot licenses after revising its sales strategy, underscoring the significant interest. However, simply purchasing a license is not a strategy.

To sustainably enhance productivity rather than jeopardise it, Swiss SMEs must adopt a structured approach. This begins with clearly defining specific business requirements. In my practice, I've often seen companies focus on a perceived "onboarding dilemma" while the actual core problem is seasonal peaks leading to massive overload. An AI tool must address precisely this core problem to be relevant. For Huber Treuhand GmbH in Thurgau, for instance, it was managing the tax season from April to September. An 'AI Tax Mentor' could relieve pressure on Sarah, the mandate manager, and better support temporary staff. Without this precise analysis, AI implementation remains a shot in the dark.

Another crucial factor is governance. Without a clear governance structure for AI initiatives, chaos, duplication of effort, and inconsistency arise. Defined responsibilities, clear roles, and a framework for AI projects must exist. This minimises risks and ensures that decisions are made in a structured manner. Kroger, a US retailer, has demonstrated how such corporate AI governance with clear roles can be successfully implemented. For Swiss SMEs, this might initially sound like over-regulation, but it is the key to safe and effective utilisation.

What specific risks does the uncontrolled use of AI tools pose to Swiss SMEs, especially on Fridays?

The uncontrolled use of AI tools carries significant risks for Swiss SMEs that extend far beyond a single day of the week. The suggestion to switch off AI on Fridays might intend to prompt employees towards critical reflection, but it misses the core issue. The real risk lies in the lack of strategy and governance, not the day of the week. One of the greatest dangers is the loss of data sovereignty and non-compliance with the Swiss Federal Act on Data Protection (DSG).

Employees who carelessly input sensitive customer data or internal business information into public AI models risk data protection violations and reputational damage. Very few public AI tools guarantee data storage and processing exclusively within Switzerland, which is an absolute must for many SMEs, particularly in the finance or healthcare sectors. The risk isn't that AI "goes rogue" on Friday, but that critical information is unknowingly disclosed every day. Only 12% of Swiss companies use machine learning, indicating a lack of understanding and perceived complexity – and thus an increased risk with uncontrolled usage.

Another risk is over-reliance on AI outputs. If employees blindly accept the output of AI tools without critical review, it can lead to errors, inconsistencies, and even legal problems. In fields like fiduciary services or legal consulting, where precision is crucial, this can have fatal consequences. Employees' abilities to think independently and solve complex problems can atrophy. This isn't a matter of the day of the week, but of corporate culture and training.

Why should Swiss SMEs restrict AI usage on certain days like Fridays, and what are the alternatives?

The idea of "restricting" AI usage on specific days is essentially an attempt to regain control and encourage employees to engage in critical reflection. The original proposal to switch off AI on Fridays aimed to enable a "digital detox" and foster human creativity and problem-solving, which could be impaired by the constant use of AI assistants. It's about creating a pause where employees can once again rely more on their own abilities and seek direct interaction with colleagues.

However, a blanket ban is a blunt instrument. It doesn't solve the problem of a lack of strategic integration and can even lead to frustration if useful tools are suddenly unavailable. An alternative is not to switch off the tools, but to manage their use purposefully and train employees. The adoption of AI tools in Swiss SMEs is seen as a strategic approach for secure AI usage, particularly through the integration of Copilot into Microsoft 365. This underscores the need for conscious implementation.

Instead of a ban, SMEs could introduce specific "AI-free zones" or "reflection times." This doesn't mean switching off the technology, but rather encouraging employees to consciously work without AI during these times, for example, to hold brainstorming sessions or solve complex problems collaboratively as a team. In my work on the 2025-2028 roadmap for AI strategy, I emphasised the importance of creating a foundation that includes a central data platform and a harmonised data architecture. This enables the controlled and targeted use of AI tools, rather than relying on ad-hoc solutions.

Huber Treuhand GmbH, with its 8 employees and over 320 active clients in the canton of Thurgau, is an excellent example of how to approach AI strategically. Instead of a blanket ban, a project to develop an 'AI Tax Mentor' is being evaluated to manage seasonal peaks. This is a concrete problem for which a specific AI solution is being sought. This is not about switching off, but about targeted, controlled addition. The team, consisting of practitioners like myself, precisely analyses the requirements to ensure that AI not only reduces workload but also enhances the quality of mandate processing.

The true alternative to "switching off AI on Fridays" is therefore a proactive, informed, and strategic use of AI. This means developing a clear strategy, identifying specific use cases that promise the greatest added value, establishing a robust governance structure, and comprehensively training employees. It's about maintaining control over the technology and using it as a targeted tool, not as an uncontrollable process.

Lukas Huber, as founder of schnellstart.ai and an experienced practitioner with an IPSO certification in AI Business, sees the future of AI usage in Swiss SMEs in such a structured approach. It's not about bans, but about empowerment – empowering employees and empowering the company to navigate digital transformation safely. A sound data platform, as might be envisioned in a 2025-2028 roadmap, forms the backbone for such a strategy by providing a central data architecture that enables secure and efficient AI use, for example, through migration to an Azure Lakehouse.

Because at the end of the day, the question isn't whether AI is switched off on Fridays, but whether it's used intelligently enough from Monday to Thursday to make Friday's work even more efficient.

Conclusion: AI usage is not an on/off switch, but a strategic decision.

• ✅ A clear AI strategy and governance are crucial for success and security.
• ✅ Uncontrolled usage poses significant risks to data protection (DSG) and productivity.
• ✅ Instead of a ban, targeted training and building a solid data infrastructure are better paths to sustainable efficiency.

Would you like to leverage AI's potential in your SME safely and strategically? We can help you develop a tailored solution that meets your requirements and complies with Swiss data protection law.

Contact us for a no-obligation initial consultation: schnellstart.ai/en/contact