The conversation around AI in media is moving fast. Debate about how the technology could deliver real value has given way to a more urgent question: who is already capturing that value, and what are they doing differently?
Simon Crownshaw, Microsoft’s industry director for media and entertainment, is seeing transformation happening in real time. However, it is taking place at the level of individual workflows rather than across whole organisations. “We’ve definitely moved past the point of asking whether the transformation promised by AI is real,” he says. “In certain workflows, AI is already delivering measurable outcomes – faster highlight creation, automatic metadata enrichment, personalisation at scale. The question isn’t whether AI works. It’s whether you’re applying it to solve the right problems.”
That distinction matters more than it might first appear. “The organisations seeing measurable impact are those focusing on solving very specific pain points first, rather than adopting an abstract AI strategy,” says Crownshaw. “They’re identifying what issues they need to overcome and how they can use AI to solve them, in a way that wasn’t possible before.”
When media teams first encountered AI tools, the natural framing was automation: repetitive tasks completed faster and at lower cost. Crownshaw thinks that framing misses the bigger picture. The more significant shift, he suggests, is the emergence of what he calls conversational operational intelligence. This is the ability to query systems in natural language and rapidly receive answers that might previously have required a data scientist, a separate dashboard, or a phone call across departments. This condenses a process that might have taken hours into something available in seconds, as well as being available to anyone who needs it. “Instead of opening multiple dashboards or asking the data team to pull a report, people can now ask questions they couldn’t easily answer before,” he says. “Why did engagement dip in the last hour? Which clips overperformed in one region versus another? What is likely to trend in the next 10 minutes? And they’re getting useful answers, in real time.”
He continues: “It’s easy to categorise the shift as just automation. But what you’re actually getting is intelligence between all the different layers. AI agents can reason across it, trigger workflows and recommend action.” For live production environments, where the stakes of a slow decision are immediately visible, that difference is especially profound.
Crownshaw identifies three areas in which early adopters are consistently seeing measurable impact. The first is reduced production friction. Live environments are unforgiving: there is no time for a team to convene around a dashboard to diagnose the problem causing a graphic to fail on screen, a network scheduling issue or unexpected latency. AI is now surfacing these anomalies earlier, often before a crew member has noticed them, and making the underlying data far easier to act on. “I’m no longer having to search frantically for something and not even know if it exists,” says Crownshaw. “Using AI, I can very quickly understand what I need, when I need it.”
The second is monetisation. “The optimisation that has occurred there is phenomenal,” says Crownshaw. “What you’re seeing is the ability to personalise streams – dynamic advertisement insertion is informed by real-time performance signals, making sure the messaging is aligned.”
The third is content production speed. Short-form content has, he explains, become substantially faster to turn around, with tagging, clipping, localisation and archiving handled in ways that free up editorial teams to focus on storytelling. “If there’s one thing that matters most in media, it’s speed,” says Crownshaw. “We’re getting there, especially for short-form content. The longer workflows are still being worked through.”
However, technology adoption is rarely the hardest part of a deployment. Changing the culture of an organisation is a necessary step in taking advantage of AI technology but poses a significant challenge in leadership and data management. Editorial, production, engineering and commercial teams must now be working from shared data models rather than separate silos, with governance embedded from the outset rather than retrofitted. “It’s moving from a tool mindset to a system mindset,” says Crownshaw. “AI will amplify silos if they aren’t addressed. If teams are pulling in different directions, AI just makes that problem more visible and more consequential.”
He points to sports leagues the NBA and the Premier League as organisations that have made this shift deliberately. Both have combined executive alignment with technical leadership and a genuine willingness to rethink process ownership from the ground up, shifting from a collection of siloed teams to something closer to a shared operational model.
The cultural and structural shifts Crownshaw describes are, in his view, precisely what Microsoft’s platform is designed to support. Where others bolt AI on as a feature, Microsoft focuses on enterprise-grade orchestration, and this is what makes the difference, he argues. Rather than layering AI over existing systems, the aim is to embed intelligence across the full stack of tools a broadcaster or production house already operates. “When you look at what we do across Microsoft Azure AI infrastructure, our Copilot framework, our agent capabilities – they have to sit across multiple systems and reason across all of it,” says Crownshaw. “And then responsible AI – how we embed the right process, the right trust, the right governance into those workflows – is core to everything we do.”
The Premier League partnership illustrates what this looks like in practice. Microsoft is building connected layers across production, performance and consumer data, enabling personalised experiences in sports video workflows and companion apps. “We can isolate less relevant content, remove irrelevant segments and give statistical overlays that make the most sense,” says Crownshaw. “That’s measurable transformation in terms of how we deliver the right outcome for a customer to the fan.”
Media ecosystems are complex, and Microsoft’s approach is built around working with the partners already embedded in them. “We can call Adobe’s services directly through our Foundry APIs to Firefly, and that directly impacts editorial workflows and post-production,” says Crownshaw. Asset management integrations with IPV and Prime Focus Technologies give AI agents direct access to media libraries, enabling smarter retrieval, transcoding decisions and metadata management at scale. Meanwhile, the combination of Microsoft and Nvidia’s infrastructure accelerates compute for rendering and AI-driven content generation. “When you put Microsoft and Nvidia together, it’s incredibly powerful,” says Crownshaw. The result is a platform where partners are no longer disconnected tools but components of a unified, orchestrated architecture.
For Crownshaw, the direction is no longer in doubt. “We’re seeing an emerging pattern,” he says. “The tools Microsoft has connected to the partner ecosystem are driving a meaningful and consistent architectural discussion about how to deploy these things at scale.”
Across the industry, he argues, the organisations that have moved from experimentation to deployment are beginning to pull away. The distance between early adopters and more cautious organisations is only likely to grow.
“The moment is real,” he says. “It opens up the ability for organisations that have been stuck in legacy applications for a very long time to transform their business in ways they’ve never done before. They can’t sit back anymore and say change isn’t possible, because with AI, it is.”
Partner perspective
Technology Record asked Microsoft partner Reply how it is using AI technology to offer new capabilities to media organisations
“The most powerful content experiences reach audiences in the moment. With Microsoft AI and Azure, Reply enables organisations to turn data into personalised engagement at scale,” says Richard Acreman, executive partner at Reply. “At Art Basel, we built an Azure AI-powered image recognition platform that identifies artworks in under two seconds, delivering real-time attribution and gallery insights worldwide. For the Vatican, we processed 400,000 photogrammetry images to create a scalable digital twin of St. Peter’s Basilica. And at the Billie Jean King Cup, our cloud-native Match Insights app (pictured) delivers real-time AI analytics that empower coaches and players to act instantly. Microsoft AI doesn’t just distribute content, it makes every interaction smarter and more meaningful.”
Discover more insights in the Spring 2026 issue of Technology Record. Don’t miss out – subscribe for free today and get future issues delivered straight to your inbox.