•
2 minute read
Cloud dependency is a risk position
The Cyber Monitoring Centre has published new research on UK exposure to cloud infrastructure failure. Companies accounting for more than 60% of UK business revenue depend on cloud services for functions critical to running the business. Among the FTSE 100, the UK's hundred largest listed companies, the figure passes 80%. Counted by number of firms alone, the share is 11%. The exposure concentrates where the economic weight is.
The research covers the UK. The structure it describes travels, and for commodity trading firms it lands on a familiar concept.
What the research measured
Dependency concentrates at two levels. Most firms that use cloud services rely on a handful of global providers, and within those providers the critical workloads gather in a small number of regions. Running more than one provider does less than it appears to, because firms tend to place their most important workloads in the same few regions, so the redundancy is thinner in practice than it looks on a diagram.
The sharpest finding sits underneath the headline numbers. Many firms cannot map the dependency at all. They cannot say which regions they rely on or how much revenue runs through them.
A position, in trading terms
A commodity trading firm treats exposure as something to be measured and limited. Counterparty, price, credit, country: each gets a number and a ceiling. Infrastructure dependency has the same shape. It is concentrated, and it sits on infrastructure someone else controls, which means upstream incidents and recovery timelines are inherited rather than managed.
An exposure a firm cannot map is an exposure it cannot limit. That is the position the research describes, and most firms are carrying it by default.
Designing the position out
For a Commodity Trade and Risk Management (CTRM) platform, the deployment model sets the dependency profile. Euclid deploys on client-owned infrastructure or inside Euclid's private Swiss data centre. Either way the firm knows where the system runs and which jurisdiction applies, and the recovery path is its own to design. The record supports the choice: Euclid runs at 99.99% uptime, against 91 to 96% for cloud-dependent platforms.
The same decision governs the intelligence layer. Euclid AI brings private AI to the operational core of commodity trading and runs inside the same environment as the platform it serves, so the AI carries the platform's dependency profile instead of adding a new one.
Continuity as a design decision
The Cyber Monitoring Centre frames its findings as a governance question, and for most industries that is the right frame. Commodity trading can go further, because the discipline this requires already exists in-house. Treat infrastructure dependency the way the firm treats any other position. Measure it, then decide how much of it deserves to exist at all.
In commodity trading, continuity is the business. A sovereign deployment keeps it in the firm's hands.
