Cyber Threats Have Shifted - Insights From Berlin

I was at the inaugural Germany-UK Business-Government Forum in Berlin on Monday.

Carolyn Esser, Chief Corporate Affairs Officer at Darktrace, said something that stopped me in my tracks.

She described two 'phases of AI' in the cyber threat landscape.

Phase 1: Large language models enabling bad actors to launch more targeted, multilingual, scaled attacks. The barrier to entry for cybercrime collapsed.

Phase 2: This is the one that should concern every B2B founder and is happening right now. Amid the pressure to adopt AI into your business, to deploy agentic AI, to automate workflows - every one of those AI agents working on your behalf is a new attack vector.

Your AI workforce is also a door your competitors and criminals can walk through.

But here's the part that gets buried in the headline statistics about enterprise breaches:

Historically, cybercriminals targeted large companies. Highest reward, most visible impact. Those companies also built the most robust defences.

In the new environment, attackers don't need to go after the big fish. They can pursue hundreds of smaller companies simultaneously, at scale, with AI doing the work for them.

If you're running a B2B tech business between €10m and €50m, you are now the preferred target. Not a side effect or collateral damage - you have become the primary target.

Another striking insight from the conference was the disproportionately high amount of British innovation fuelling cyber and AI in Germany right now, as well as the new investments by companies like Boehringer Ingelheim to tap the massive AI talent pool in and around Kings Cross in London.

The political framework for UK-Germany industrial cooperation is stronger than it has ever been and the commercial opportunity is real and time-sensitive. However, the infrastructure you build your growth on needs to be secure.

This isn't a technology problem. It's a business risk problem. And it belongs on the agenda at the top table in your business.