{"id":61060,"date":"2025-10-10T16:45:59","date_gmt":"2025-10-10T14:45:59","guid":{"rendered":"https:\/\/www.secuinfra.com\/?p=61060"},"modified":"2025-10-14T13:13:50","modified_gmt":"2025-10-14T11:13:50","slug":"world-summit-ai-2025-between-innovation-frenzy-and-security-reality","status":"publish","type":"post","link":"https:\/\/www.secuinfra.com\/en\/news\/world-summit-ai-2025-between-innovation-frenzy-and-security-reality\/","title":{"rendered":"World Summit AI 2025: Between innovation frenzy and security reality"},"content":{"rendered":"\n

Amsterdam, October 8-9, 2025 – Two days, one stage for the future: The World Summit AI 2025<\/strong> has once again positioned itself as a global hotspot for developments in artificial intelligence – and we at SECUINFRA were right in the middle of it. With a clear goal: to place the innovation dynamics of the AI world in the context of cyber defense<\/strong>. After all, what use is a technological quantum leap if security falls by the wayside? <\/p>\n\n

The who’s who of the AI industry came together at Taets Art & Event Park<\/strong>: from giants such as AWS<\/strong>, Google<\/strong>, SAP<\/strong>, Oracle<\/strong> and Cloudflare<\/strong> to visionary start-ups such as DeepL<\/strong> – the diversity of players was impressive. It was all the more remarkable who was not represented: neither Microsoft nor central AI pioneers such as OpenAI, Anthropic or Meta were present. Technical enablers such as NVIDIA and AMD were also surprisingly absent – a gap that was noticeable in the midst of the AI hype. <\/p>\n\n

Knowledge in abundance – but not everyone got in<\/h3>\n\n

The program? Tightly timed, curated, full of impulses: “Deep Dive Tech Talks”,<\/strong> “Innovation Insights”,<\/strong> interactive workshops<\/strong> – an impressive density of formats, often in parallel, which made selection a strategic discipline. On the first day in particular, the event reached its spatial limits. Overcrowded halls, no more admission – the interest in AI has long been more than just hype, it is reality. Day two was more relaxed, without losing any of the depth of content. <\/p>\n\n

Lecture highlights: Between sharp intellect and tangible added value<\/h3>\n\n

We were particularly inspired by the presentation “Use Case: Lies, damn lies and GenAI”<\/strong> by Jon McLoone (Wolfram Research Europe Ltd.)<\/strong> – a rhetorically brilliant digression on the interactions between truth, deception and generative AI. A presentation that not only provided knowledge, but also encouraged critical thinking – a central claim that we also live by in our daily work. And McLoone? Approachable, open, in direct dialog with the audience – a real added value. <\/p>\n\n

Equally valuable: Arthur Eyckerman’s (Elastic)<\/strong> presentation “Ease Into AI: Revolutionizing Security Operations with Elastic and AWS”<\/strong> at the AWS stand. Here, our colleague Aron was able to go into more depth afterwards – an opportunity that is worth its weight in gold at such events. <\/p>\n\n

Buzzwords, vibes and a clear message: no progress without safety<\/h3>\n\n

The same themes echoed through all the sessions: AI agents<\/strong> are the new standard. Security and guardrails<\/strong> are the new mandatory fields. And in the middle of it all, terms such as “vibe coding<\/strong> ” – fascinating, new, with enormous potential, but also with risks that cannot even be fully assessed yet. <\/p>\n\n

The realization: AI is not dangerous per se – but its use without control is.<\/strong> With autonomous AI agents in particular, it is clear that where machines make decisions, humans must determine the framework conditions. Security architecture becomes both an enabler and a guardian.<\/strong><\/p>\n\n

The shadow of innovation: AI in the service of attackers<\/h3>\n\n

Even if it wasn’t in the spotlight, it was clear in many conversations that cyber criminals have long since operationalized AI.<\/strong> They are faster, more adaptive and more targeted. Traditional defense strategies no longer work. The game has changed and we know the new rules.<\/strong><\/p>\n\n

For us as cyber defence specialists, it is clear that anyone who wants to protect organizations today must understand how modern attackers think, what tools they use and how we can counter them both technically and strategically<\/strong>.<\/p>\n\n

Why AI is not an option for SECUINFRA, but a mission<\/h3>\n\n

Our participation in the World Summit AI was not a trip to the Silicon Valley circus tent – it was a strategic deep dive<\/strong> into the future of IT-supported work. After all, our customers will increasingly rely on AI in the coming months and years – be it to automate entire processes<\/strong> or to provide intelligent support for human decisions<\/strong>. <\/p>\n\n

But: AI needs access to data, to systems, to decision-making structures.<\/strong> And this is precisely where cyber security comes into play. Anyone who fails to think about this opens the door to uncontrolled risks. <\/p>\n\n

That is why we are already investing in:<\/p>\n\n