Europe's AI caution creates security vulnerability, analyst warns

European organizations face a paradox: skepticism toward artificial intelligence coexists with mounting threats from AI-powered cyberattacks. To protect infrastructure and keep pace with American technology companies, the region must fundamentally shift how it approaches AI deployment, argues Holger Mueller, vice president and principal analyst at Constellation Research Inc. Mueller contends that human-led defenses cannot effectively counter threats powered by artificial intelligence.

Mueller shared his perspective during Oracle's "AI Cyberattacks Are Escalating: How to Secure Your Data Now" event, speaking with Dave Vellante and Krista Case, both principal analysts for theCUBE Research. The conversation centered on securing enterprises against AI-driven threats and Oracle Corp.'s agentic AI capabilities.

Skepticism and sovereignty concerns slow European adoption

Europe is traditionally always skeptical of innovation coming from the U.S. With AI, there's a lot of conservatism, a lot of waiting. Sovereign cloud is a big topic, but it's also a big excuse. It's not a good idea waiting for sovereign cloud when you're doing business on a global level, like many European countries are doing.

Holger Mueller, Constellation Research Inc.

Mueller acknowledges that people inherently distrust automation. Yet paradoxically, deploying additional AI systems may represent the path forward for closing security vulnerabilities in agentic AI frameworks. The Hugging Face security incident exemplifies this dynamic: when an autonomous AI agent executed an intrusion, defenders subsequently relied on an open-weight model to examine the breach after commercial frontier models declined to process the forensic data.

Technical debt and infrastructure challenges

They need to resolve the tech debt they accumulated over 20 years as fast as possible because AI really makes you pay for tech debt. You may not have managed, may not have experienced on-premises [solutions]. But now you can only consume pretty much all of AI from an elasticity perspective.

Holger Mueller, Constellation Research Inc.

Constellation Research recently published findings on "agentic AI washing," describing instances where vendors overstate the degree to which agentic AI has been integrated into their offerings. When evaluating Oracle's AI database portfolio against Constellation's rigorous benchmarks, the firm determined that Oracle delivered genuine agentic AI functionality without engaging in such exaggeration.

The evolution of agentic AI priorities

The agentic AI uptake last year was for the advanced companies to create the data source, the data plane for those agents. This year, everybody's fighting about what is the better agentic framework. Next year is going to be the big renaissance of scaling the backend systems. And this is a great message for Oracle customers knowing that they have a highly scalable database.

Holger Mueller, Constellation Research Inc.

Oracle's long tenure as a provider of secure transactional databases positions it favorably as enterprises navigate agentic AI deployment. The company's strategy—offering traditional database capabilities alongside options to migrate external workloads to a data lakehouse—addresses a critical requirement: database scalability underpins agentic success, allowing organizations to transition to cloud infrastructure while preserving agent functionality. Security enhancements must accompany each phase of this transition.

https://www.youtube.com/embed/_B9G0AmVbwA?feature=oembed

You need to have a selection in saying which models you use and you need to have a playground to figure that out. You need to band up with a few CISOs to do red team, not red hat, but black hat, white hat attacks or whatever [you need] to figure out what is going on there. One side point on this … is if you think state actors will spy on you, you need to be quantum safe.

Holger Mueller, Constellation Research Inc.