16 SEP 2026
InfraXMedia
The Xcelerated Compute Show London 2026The Compute Infrastructure Questions Every AI Buyer Should Ask
· London, United Kingdom
John Bradshaw · Field CTO, Akamai · Speaker · Writer
Most digital transformations succeed at the technology and fail at the business.
- $300M cloud transformation led
- 8→150 team built in 18 months
- $2M/qtr FinOps savings delivered
John Bradshaw is a cloud and AI strategist at Akamai Technologies, focused on cloud infrastructure, edge computing, and AI inference. He speaks at conferences across Europe and the Middle East — from WeAreDevelopers World Congress to The AI Summit London — on cloud economics, API security, and bringing AI to the edge.
Explore by topic
All topics → Cloud Economics & Sovereignty Cloud bills rarely blow up on compute — they blow up on egress, lock-in, and architecture decisions nobody revisits. Here's where I work through what actually drives the spend, when "just put it in the cloud" stops being the cheap answer, and how to keep control of your own stack. Edge & Distributed Architecture The interesting compute is moving away from a handful of mega-regions and out toward the user. This is my thinking on edge and distributed architecture — why latency, data gravity, and sovereignty keep pushing workloads outward, and what that changes about how you build. AI Foundations & Adoption Most of what gets called an AI strategy is really an infrastructure decision wearing a different hat. This is the broad end of my AI thinking — what the models actually are, where the compute has to sit, what enterprises keep getting wrong on the way in, and the running commentary I've given on all of it. AI Inference Economics Training gets the headlines; inference gets the bill. Here's where I dig into the economics of running AI in production — when an API stops being cheaper than self-hosting, where inference should physically live, and how to stop the per-query cost from eating the margin. Agentic AI & MCP Chatbots were the demo; autonomous agents that take real actions are the product. This is my thinking on the patterns that hold up in production — what to let an agent do on its own, where to keep a human in the loop, and how tools and protocols like MCP make it work at scale. Security & AI Governance Agentic AI changes the threat model: the caller is no longer a person, and the guardrails can't live in a PDF. Here's where I work through API security when the caller is an agent, zero trust in practice, and governance as an operating model — controls that trigger, gates where decisions are irreversible, and audit trails that survive contact with a model. Digital Transformation Most digital transformations succeed at the technology and fail at the business. Here's where I look at what actually moves a programme — the operating model, the funding, the decisions you can't delegate — and why the platform choices are usually the easy part. Media: Content to Experience The streaming wars were won on content; the next ones will be won on experience. Here's where I look at what that shift means for media — why discovery, personalisation, and real-time delivery now matter more than the size of the library.
Upcoming Talks
View all →Media
View all media →AS Watson (a member of CK Hutchison) named 31 global technology partners at its Technology Partner Day 2026 to support the next phase of its AI-powered O+O (offline-plus-online) retail transformation — spanning enterprise productivity, personalisation, cybersecurity, and digital commerce. John Bradshaw is cited among the partner contributors.
A video interview on how enterprises scaling AI deployments are straining traditional cloud infrastructure, and why an "AI Grid" model — distributing AI inference across edge locations rather than concentrating it in a handful of facilities — changes the economics of running AI in production for UK and international organisations.
Entrevista con John Bradshaw durante su visita a España, donde analiza el cambio de paradigma que supone el estallido de la IA: cómo el edge computing y la IA agéntica transforman la relación entre empresas y clientes, por qué los modelos especializados reducen costes frente a los grandes LLM, y por qué «la inteligencia artificial es tan buena, o tan mala, como los datos que la alimentan».
Writing
View all writing →AI governance fails when it is a policy without decision rights. A practical operating model for lifecycle gates, accountable roles, enforcement and time-bound exceptions.
A practical control map for deploying Model Context Protocol safely: identity, permissions, supply chain, approval gates, logging and operational ownership.
The practical AI FinOps metrics that connect token, GPU, retrieval and agent cost to accountable operational and business decisions.