Several major transitions are reshaping the media industry at once. That was my strongest impression from IBC 2026. Audience habits, distribution platforms, and technology infrastructure are changing, alongside growing cost pressures and rapid advances in AI. For me, this raises a practical question: how can technology help broadcasters adapt without repeatedly rebuilding the way they work?
Broadcast technology must make change easier
Broadcasters need to launch services, test distribution models, and adapt workflows without another multi-year transformation programme every time their priorities change. Flexibility is becoming a basic requirement. The economics matter just as much: the question is not only what a system can do, but what it costs to implement, integrate, and adapt over time.
The move from dedicated hardware to software and cloud is part of this transition. But hosting an application on a remote server is not enough. Systems need to scale with demand, work with other tools, and give customers access to their own operational data. An integrated environment should not be a closed one.
Where AI agents fit into connected broadcast workflows
Transcription, subtitling, and metadata processing have already shown the value of automating individual tasks. At IBC 2026 the next step was visible: working, deployed agents carrying out sequences of actions on their own. Almost all were built around systems, automating functions inside an application. I saw very little built around a process.
Moving data between systems is not the problem; interfaces have done that for thirty years. When a programme moves in the schedule, someone still has to work out which promos now point at the wrong slot and what changes in the transmission plan — conclusions drawn from the change, entered by hand. The same gap opens between two processes sharing one database, which is why systems integration never closed it.
Where that handover is algorithmic, rules and workflow have automated it for years. What remains is contextual: understanding a situation, weighing what it means for several things at once, and deciding. That is where an agent belongs — following the process rather than the systems architecture, and working with many data objects across several systems at once. Rules and interfaces remain better at everything they already cover.
For smaller media organisations, the goal need not be primarily about reducing headcount. It could mean managing more channels, more content, and more distribution platforms with the same team. Technology should be judged by the work it makes possible, not only the work it removes.
Connected broadcast workflows need structured data
There is an important difference between an impressive demonstration and dependable day-to-day operations. Greater agent autonomy requires clear business rules, permissions, and a gradual progression from recommendations to actions carried out under human supervision.
Above all, it requires reliable, structured information about content, rights, contracts, schedules, and what actually aired. Without that foundation, automating decisions risks multiplying inconsistencies rather than resolving them. AI does not remove the need for well-managed operational data. It makes that need more pressing.
From connected information to connected operations
In the 1990s, integrated broadcast management systems helped connect information scattered across departments, spreadsheets, and separate databases. That transition integrated the information. Connected broadcast workflows could extend that integration to the process itself — the actions and decisions that still depend on a person reading the outcome of one step and working out what the next one should be.
That is my main takeaway from IBC 2026. The question is not simply which features software will offer next, but how it will help media organisations change the way they work. In a market whose future shape remains uncertain, the ability to adapt may be the most valuable capability of all.
Josef Vašica, CEO of PROVYS Technologies
