The Business of AI

Most of us think we are better at AI than we really are, by almost 20 points on average. In this episode, Ozan, founder of AISA, an AI fluency assessment platform, explains why AI fluency has become a core workforce skill and how to measure it properly. Drawing on a background in sociology, psychology and neuroscience, he sets out a five-part framework: prompting and communication, critical thinking, technical understanding, workflow and application, and safety and responsibility, including the judgement to know when not to use AI at all. He explains why dull multiple-choice tests are being replaced by a playful, conversational assessment full of games, which captures far richer data and lets organisations see fluency team by team and spot exactly where the gaps are.

Ozan also tackles the human side of adoption: how to tell the enthusiasts from the sceptics, why one-size-fits-all training wastes money, and how curiosity, shareable certificates and a little gamification bring reluctant staff on board. He reveals that 65% of people think AI skills should count in hiring decisions, yet senior leaders are often the most nervous about having their own scores seen. His biggest worry is leaders making decisions about jobs and workflows without a real baseline of their own AI fluency. The good news? It is still day one, and the gap can be closed in a matter of months.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

Quantum computing is not a distant dream: you can run real workloads on it right now. Richard, co-founder of ORCA Computing, did his PhD in quantum physics 25 years ago, when nobody imagined commercialising it, and later ran the UK’s pioneering quantum technology programme at Innovate UK. In this episode he demystifies the basics: qubits that can be one and zero at the same time, entanglement, and why quantum computers can explore many possible answers in parallel. He explains what really matters when comparing machines, from the number of qubits to error rates, speed and connectivity, and why using photons travelling through optical fibre at room temperature is such a powerful approach.

Richard then turns to the question on everyone’s lips: what can quantum do for AI? Today’s quantum computers are too small to run whole models, but used as accelerators in hybrid systems they can power the early layers of generative AI models. In molecular chemistry, this is already producing novel molecules and peptides that classical models alone miss. He discusses the UK’s early lead, the funding race against well-capitalised US rivals, and how businesses can access quantum systems hosted alongside GPUs in a London data centre today, without needing to understand any of the physics.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

AI adoption is accelerating, but where is the return on investment? Too many businesses are investing in AI without defining the problem they want to solve, leading to expensive pilots, fragmented systems and productivity gains that never reach the bottom line. As agentic AI introduces new risks and complexity, the next competitive advantage won't come from deploying more tools. It will come from delivering measurable business outcomes while maintaining security, trust and accountability.
Sonia, former Head of Business AI at SAP UK and Ireland and founder of Nashmind Consulting, explains why businesses must move beyond the AI gold rush towards an era of accountability. She explores how leaders can identify trustworthy AI vendors, avoid costly implementation mistakes and establish governance frameworks before autonomous agents become unmanageable. From cybersecurity and executive responsibility to the UK's evolving approach to AI regulation, the discussion offers practical lessons for businesses navigating AI transformation. Successful AI adoption isn't about moving fastest. It's about choosing the right problems, asking tougher questions and making AI deliver real commercial value.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

As companies race to deploy increasingly autonomous AI agents, the biggest threat isn't just technical failure. It's the growing gap between AI capabilities and the governance, accountability and safeguards needed to control them. From cybersecurity and unintended agent behaviour to reputational damage and regulatory exposure, AI risk is becoming a boardroom issue. Yet many organisations still treat it as an IT responsibility rather than a strategic business priority.
In this episode, Giovanni, an AI risk and governance consultant and founder of Quantidal, explores how businesses can build practical risk frameworks, establish clear accountability and prepare for the evolving EU AI Act, including its implications for UK companies. The conversation tackles the emerging dangers of agentic AI, the challenge of balancing autonomy with control, and the ethical questions surrounding artificial superintelligence. With global competition accelerating AI development, the central challenge for business leaders is clear: how do you maintain competitive advantage without losing control of the technology driving it?

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

Most companies are still treating AI like software and that may be why so many struggle to turn promising pilots into enterprise value. The shift to agentic AI changes the equation: autonomous agents can execute work, make decisions and pursue objectives, which means organisations need to think less about deploying tools and more about redesigning how work gets done. That brings operating models, accountability, governance and workforce strategy firmly into the AI conversation.

Ray Welsh, an enterprise technology veteran now focused on agentic workforce strategy, argues that AI agents should increasingly be managed as a new class of worker. The competitive advantage will not come from access to AI alone, but from how effectively businesses orchestrate human and digital labour. Leaders need clear decision rights, stronger governance and employees capable of supervising, challenging and collaborating with AI, turning agentic adoption from a technology project into a business transformation.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

AI adoption is soaring, so why are half of users saving no time at all? Deloitte’s new GenAI Workforce survey of 25,000 UK workers exposes a growing gap between access to generative AI and meaningful business impact. While 63% of workers have used AI at work, only 12% use it multiple times a day. The result is a workforce of AI “dabblers”, where headline adoption figures can mask limited changes to how work actually gets done.

Ben Stanton, argues the productivity problem isn’t primarily the technology, it’s the infrastructure around it. Half of AI users report receiving no training, only 28% say their organisation has a clear and accessible AI policy, and 65% say leaders aren’t communicating effectively enough about AI. The opportunity for businesses is therefore less about buying another tool and more about creating the leadership, training and confidence that turns experimentation into genuine productivity.

Access the report here: https://www.deloitte.com/uk/en/issues/generative-ai/genai-workforce-survey.html

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

Are AI data centres really about to consume Britain’s power and water, or are the headline numbers hiding a very different story? Spencer argues that some of the biggest assumptions about AI infrastructure need challenging. While around 80GW of data-centre power requests have reportedly entered the UK connection queue, he estimates genuine AI-related demand could be closer to 10GW over the next five to ten years. He also explains why UK cooling systems can look very different from the water-intensive facilities behind some US headlines.

But this isn’t simply an infrastructure debate. Spencer estimates that building 10GW of capacity could require £150 billion of investment, while creating demand for engineers, electricians, plumbers, construction workers and other skilled roles. With data centres increasingly able to operate beyond London, AI infrastructure could become a catalyst for regional regeneration if Britain can secure the power, capital, planning and public confidence required to build it.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

What happens to consulting when AI can do in days what teams of consultants used to spend months delivering? Neil argues that AI is rapidly collapsing the cost and time of traditional software and services, forcing businesses to rethink where value actually sits. Instead of starting with models, agents or productivity targets, companies should focus relentlessly on measurable outcomes: revenue growth, lower costs and reduced risk. The winning model is “compound interest AI”: deploy small, high-value workflows quickly, prove their impact, then build momentum through repeated delivery.

The bigger shift is that building the technology itself is becoming less valuable. Differentiation increasingly comes from proprietary workflows, data preparation, carefully refined AI instructions and continuous human oversight. Neil’s approach at Neologik combines reusable industry blueprints with customer-specific configuration and a model-agnostic platform pointing towards a future where businesses buy outcomes rather than armies of consultants or long software projects.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

This week Tim gets a ride at the buzzy Uber x Wayve robotaxi launch in London ("AI on wheels" and, "a real milestone for the British AI industry"), digs into the £100m government fund backing British AI startups, unpacks the Matt Clifford-to-Anthropic conflict-of-interest row, pushes back on calls for a data centre moratorium, and covers Nvidia's $12.9bn Hugging Face buy, new AI "kill switch" bills in the UK and US, and Matt Lucas and Hugh Bonneville's campaign to protect actors' voices from cloning. A quick look at the week in British AI.

The views expressed in this video are Tim's own personal opinions and do not necessarily reflect the views of UKAI or any of its members.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co

Why does AI-generated code so often look sloppy, even when it works? Kevin Coyle, founder of Drutek, joins the podcast to explain why the missing ingredient isn't better prompts, it's context. He breaks down how design systems give AI tools the grounding to produce consistent, on-brand results, and shares his "harvest" framework for getting every employee, not just developers, involved in building with AI.

The conversation then turns to the UK's competitiveness, with Kevin drawing on his experience running a US office to compare tax, talent and ambition on both sides of the Atlantic. He's blunt about the barriers facing British businesses, but optimistic that the UK's foundations are strong enough to compete, if policy catches up with the moment.

AI is our Business.

UKAI is the Trade Association for AI businesses across the UK. Join us, ukai.co