Practical analysis on NIS-2, ISMS, AI governance and mid-market cybersecurity. From 25 years of compliance practice and from OdySecure, the security platform by Cybervize.
A CEO says "Security? We've implemented it." Three questions later, silence. Why cybersecurity without ongoing cadence fails, and which four routines ensure real sovereignty.
The most common AI mistake in companies is not a prompt engineering problem, but it is the unreflective copy-paste reflex. Why data masking is the crucial safety mechanism for AI usage.
The Clinejection case demonstrates how prompt injection via GitHub Issues can manipulate AI agents to inject malicious code into release workflows. Automation without security-by-design creates dangerous new attack vectors.
Many companies treat NIS2 as a tick-box exercise. But compliance is not the same as resilience. The Cross-Border Cybersecurity Tour #2 in Saarbrücken made it clear: a functioning security operation outweighs any tool collection.
NIS-2 does not fail at technical gaps. It fails at unresolved ownership. What it means to anchor responsibility concretely.
70% of SMEs treat NIS2 as a compliance checkbox. But organizations that see it as a strategic lever can turn regulatory requirements into operational excellence and genuine resilience.
When 'everyone and no one' is responsible for NIS-2, implementation fails before it starts. Why ownership is the underestimated success factor and how a structured assessment creates clarity.
Humans are not getting worse at cybersecurity. AI is getting better at finding their mistakes. This fundamentally changes the rules of the game and makes resilient systems the most critical response.
Alexander Busse speaks at the CROSSBORDER CYBERSECURITY TOUR #2 in Saarbrücken on how NIS2 compliance can drive operational excellence. Why 70% of SMEs misjudge the regulation and how to turn it into a genuine competitive advantage.
OpenAI reveals: prompt injection attacks succeed despite protective mechanisms in 50 percent of cases. Why this is not a technical but a governance problem, and which five principles organizations should implement now.
Many mid-sized companies commit to Zero Trust until it becomes inconvenient. The real test does not happen in the concept document but in the permissions: Who has admin access, and why?
The attack on McKinsey's AI platform Lilli marks a new era: AI is no longer just a tool for attackers. It is the attacker itself. What this means for IT decision-makers in mid-market companies.