Artificial intelligence opens up enormous opportunities – and creates new risks. For many small and medium-sized enterprises, the market for security and support services in the field of AI cybersecurity is difficult to navigate. With the project "Cybersecurity with and for AI," funded by the Baden-Württemberg Ministry of Economic Affairs, Labor, and Tourism, the AI Alliance Baden-Württemberg is working with strong partners to create quality-assured guidance: support services are systematically reviewed, evaluated, and prepared in a practical manner for SMEs.
Artificial intelligence is changing business models, value chains, and decision-making processes at a speed that poses structural challenges for many organizations. At the same time, the threat situation in cyberspace is becoming more acute: attacks are becoming more professional, more automated, and increasingly AI-supported. Those who introduce AI not only expand their innovation potential, but also their own vulnerability.
Cybersecurity with and for AI: Why guidance is now becoming a strategic resource
Against this backdrop, it becomes clear that AI and cybersecurity cannot be considered separately. It is not just a matter of protecting traditional IT infrastructures, but also of addressing issues such as model manipulation, data poisoning, prompt injection, the secure use of generative systems, and the resilience of AI-supported decision-making processes. The quality of the technologies used is just as crucial as the quality of the accompanying support services. After all, we can only implement and control what we truly understand.
This is precisely where the project "Cybersecurity with and for AI: Reviewing, evaluating, and highlighting support services" comes in. The idea: structure, evaluate, curate. Instead of reinventing the wheel, the project team filters the market. Around 20 offerings are screened in each category, such as compliance, incident response, or AI-specific security. The end result is three to five crystal-clear recommendations per area that are specifically tailored to the needs and resources of SMEs.
Small and medium-sized enterprises (SMEs) in particular face a fundamental question of orientation: Which offerings on the market are relevant, reputable, and actually feasible? And which ones promise more than they can deliver in practice?
When it comes to support services—such as training and qualification programs or initiatives to raise employee awareness of AI-specific risks—the market is flooded with tools and promises. This situation often leads to paralysis rather than action. "Companies don't need another endless list of possibilities. They need reliable guidance," emphasizes Dr. Jan Zipp, who is playing a key role in driving the project forward within the AI Alliance.
Quality over quantity: Curated security "Made in BW"
"We are also creating a quality-assured selection from practice for practice in this increasingly relevant field of AI development," says Zipp. The aim is not to create a complete market atlas, but rather a quality-assured, action-oriented selection that provides companies with a reliable basis for decision-making.
The methodological complexity lies less in the technical examination of individual products than in the structured comparability of heterogeneous formats. This is precisely where the curatorial function of the project comes in: complexity is reduced without resorting to simplification.
The results are published using a rolling release process, so companies do not have to wait for a final report and can benefit from the findings immediately.
Safety as a prerequisite for innovation
The debate surrounding AI is often framed in terms of "opportunities vs. risks." In reality, the two sides are inextricably linked. Without robust security architectures and reliable support services, AI will not be able to realize its innovation potential—especially in small and medium-sized enterprises.
The AI Alliance is already working on topics such as the responsible use of AI, regulatory developments (e.g., in the context of the European AI Act), training, and the establishment of quality-assured information and support structures. The new project adds a key dimension to this portfolio: the systematic combination of innovation promotion and security expertise.
Like all AI Alliance projects, this one is not an isolated endeavor, but is consistently embedded in strategic fields of action. The Alliance sees itself as a transfer and structural actor between research, business, and politics—with the aim of systematically bringing together innovation and responsibility.
“Using AI responsibly, managing risks confidently”
"If we want to use the opportunities offered by AI responsibly, we must manage the risks confidently. The right support services provide the necessary expertise and strengthen the organizational culture in terms of a profitable and, indeed, secure transformation," explains Dr. Jan Zipp.
Cybersecurity is therefore not a downstream protective measure, but a prerequisite for strategic action. Anyone who uses AI needs guidance, expertise, and trustworthy structures. With its new project, the AI Alliance Baden-Württemberg is creating a solid foundation for precisely this: quality-assured transparency in a field that is increasingly crucial for the competitiveness and resilience of companies.
The ecosystem as an impact accelerator
The strength of this project lies in the cooperative DNA of the AI Alliance. Thanks to close cooperation with regional partners, the needs of companies are fed directly back into the project. The result is a common quality framework that will continue to exist beyond the one-year project period.
This new "security compass" will be rolled out gradually starting in spring 2026. For companies in the state, this means less worry about the "AI jungle" and more focus on what they do best—driving innovation.