Archive
All opinions
4,384 pieces · page 3 of 147
October 2026
-
The next passport UK exporters need may belong to their products
The EU’s product passport registry is live. British exporters should start with accountable product records, supplier evidence and a practical pilot before committing to a platform.
-
Britain may be watching the wrong quantum race
Britain should assess quantum sensing against real engineering problems. American deployment targets and British development projects justify investigation, but funding and deadlines cannot substitute for proven field performance.
-
The first autonomous ships that matter to Britain may never cross an ocean
Britain should prioritise defined maritime jobs over autonomous ocean crossings. Singapore’s port feeder programme shows where technology suppliers could contribute, and what buyers must prove before committing.
-
The future data-centre customer may need to arrive with its own power station
The next data-centre contract may depend as much on electricity infrastructure as connectivity. Britain should require credible power plans while holding developers and network companies accountable for delivery.
-
6G may matter far more to machines than it ever does to smartphone users
South Korea’s 6G ambitions and factory-focused evaluation work suggest an industrial opportunity. British suppliers should prepare, while buyers should insist on measurable benefits over existing networks.
-
Britain should learn from Germany’s industrial approach to private 5G
Germany’s local spectrum and network identifiers offer British industry a practical lesson. Judge private 5G by the production problem it solves, the full operating cost and who takes responsibility when it fails.
-
Britain’s ports should automate for better trade and better work
Britain should back port automation, with investment judged on cargo reliability and credible workforce transitions. Singapore and Felixstowe show why the decision already extends beyond buying driverless trucks.
-
Britain needs to prepare for AI workers
British companies should prepare for AI agents with bounded authority. UAE ambitions and Singapore’s governance framework point towards practical decisions about ownership, permissions, spending and recovery.
-
Post-quantum migration could become Britain’s next Y2K-scale IT programme
Post-quantum migration could become a vast, largely invisible IT programme. British businesses should fund discovery and supplier coordination now, while separating routine upgrades from systems that genuinely need replacement.
-
Buying Wi-Fi 6E in 2027 needs a clear business case
Wi-Fi 6E deserves consideration when extending a supported network. For a fresh installation, compare Wi-Fi 7 first and make the decision on installed cost, actual devices and ongoing support.
-
Quantum risk starts with the data your business needs to keep secret
Encrypted information captured today could become readable later. UK businesses should identify long-lived secrets, establish supplier responsibilities and plan supported upgrades before committing to a costly migration.
-
Satellite-to-phone backup should start with keeping staff reachable
Satellite-to-phone deserves a place in UK business continuity trials, with a defined task and a measurable result. Coverage figures alone cannot establish whether staff can communicate when it matters.
-
LiFi belongs in selected UK rooms before it replaces Wi-Fi
LiFi deserves a measured trial in selected rooms and industrial links. For UK buyers, device compatibility, installation costs and dependable support matter more than theoretical speed.
-
UK Smart Data should make software easier to leave
Smart Data could make software easier to leave, but powers to create schemes are not a complete migration service. UK businesses should demand tested transfers, clear responsibilities and priced exit work today.
-
UK AI adoption programmes give technology partners a delivery opportunity
UK AI adoption support gives technology partners an opening to deliver practical implementation. Small businesses should demand measurable pilots, clear responsibilities and full costs before committing.
-
UK businesses should start moving beyond PDF invoices
UK businesses should prepare for structured invoicing now, starting with cleaner data and a tested customer connection. Replacing a working accounting platform should depend on requirements and measured results.
-
AI Growth Zones should change your hosting shortlist before your postcode
AI Growth Zones could broaden the UK hosting market. Smaller companies should use them to widen supplier shortlists, then judge actual capacity, support responsibilities and total cost.
-
Smaller UK AI companies need usable compute, not just bigger capacity
Smaller UK AI companies can benefit from AIRR when they have a prepared experiment and a plan for deployment. National capacity matters, but usable access, engineering time and the full project budget decide its practical value.
-
UK manufacturing AI investment needs a factory business case
Government AI support should help manufacturers test a business case. Our view on funding announcements, Scan, Pilot, Scale, supplier selection and the costs of keeping a system running.
-
Choosing an AI consultant starts with making their claims testable
Choose an AI consultant by the decision they will resolve and the evidence they will deliver. Compare scope, operating costs and responsibilities before committing to a wider programme.
-
AI coding agents must earn their savings after review
AI coding agents should earn their savings after review, testing and rework. UK software firms need to measure the cost of accepted changes, not the speed at which an agent produces a patch.
-
Proxmox VE can cut private AI costs if the hardware earns its keep
Proxmox VE can underpin private AI on existing servers, but savings depend on hardware suitability, workload performance and paid-for operating responsibility. Start with a measured pilot before committing to production.
-
UK businesses should prove the workload before buying NVIDIA DGX
Rent a representative workload before committing to rack-mounted NVIDIA DGX. Ownership needs measured demand, suitable facilities and a costed operating model, not an AI ambition alone.
-
When UK businesses need an AI workflow specialist
Staff should have room to build useful automations. Specialist help becomes necessary when a business cannot reliably test, monitor or reverse the consequential actions those workflows take.
-
Mistral deserves a place on the UK private AI shortlist
Mistral deserves a UK shortlist for private business applications. Its deployment choices matter, but the decision should turn on tested answers, controlled access and the full cost of running the service.
-
When UK businesses should build with Amazon Bedrock
Build with Amazon Bedrock when a valuable workflow needs a custom application and you can fund its upkeep. Trial purchased assistants first, then judge the remaining gap against the full cost of ownership.
-
Amazon Bedrock in London needs more than a regional setting
Amazon Bedrock in London is a credible candidate for tighter regional control, but the endpoint alone proves little. The buying decision depends on the exact model, processing route, supporting services and operational ownership.
-
When British businesses should run gpt-oss locally
Local AI needs a business case beyond keeping data nearby. Compare the Cloud argues for a bounded workload, tested output quality and a funded operational owner before British businesses commit to gpt-oss.
September 2026
-
Copilot Chat is enough until the work proves otherwise
Copilot Chat can be a sensible starting point for UK businesses. Paid licences should follow evidence that broader work context or specific app capabilities improve completed work enough to justify the cost.
-
When UK businesses should pay for AI beyond Google Workspace
A separate AI assistant needs a specific job and measurable value. UK businesses should test their existing Gemini entitlement, then fund additional seats where better results or required controls justify the cost.