Amagi's first APAC edition of the AIRTIME webinar series brought together Baskar Subramanian, Managing Director and CEO, and Jay Ganesan, SVP for the APAC region, for a wide-ranging conversation on how agentic AI is reshaping broadcast and streaming workflows: from scheduling and metadata to content discovery and personalized channels.
Don't have time to read everything? TLDR:
- Agentic AI is the next major industry transition — comparable in scale to the shift from analog to digital or SD to IP.
- For every $1 spent on technology, the industry spends $2–$4 on human toil; AI is the lever to change that ratio.
- Agentic use cases span the entire workflow today — scheduling, metadata, monitoring, promo creation, and social publishing.
- Personalized channels are a reality today, not a future concept — Amagi has been running them for 18+ months.
Let's dive into the highlights. And if you want to watch the full webinar, you can right here.
What is agentic AI in media operations?
Ganesan: Agentic is probably the next biggest transition that we as an industry are going towards — moving from, let's say, analog to digital, or HD to IP. Agentic is going to fundamentally challenge the way we've been looking at the industry. All of us in our roles are going to become agent managers, defining the workflows where the agents are going to be performing most of the chore tasks, while we as humans will graduate more and more to do thinking and decision-making jobs.
From a single supply chain of content, we have increasingly moved into a multinodal supply chain. If you do have a piece of content — live or recorded — you're now talking about looking at the lifetime value of that content, starting from maybe a broadcast, a subscription model, or AVOD, all the way through to social media and FAST. Agentic would definitely play an extremely important role as we get our media operations ready for this multinodal supply chain.
How does an agentic AI scheduler work in media operations?
Subramanian: One of the jobs in scheduling is really filling that 24/7 grid, which is extremely hard. What agents are helping you do is — it understands our genre, it understands the domain, it understands what scheduling is all about, it understands all the terminology and the context that you have. We are now able to run these machine activities, fill holes, get your fillers done, and schedule them for all the off-hours. The key decisions are with the human, but the whole capability is driven by the agent.
I'm imagining a daily experience here where, in the early mornings, the system is actually working — we are not working. In the morning at 7am, you're getting a brief from the scheduler saying: I've done it, I have some decisions that I want you to make, please make those decisions for me. You're able to come back and review the grid, give some comments, and the system goes back and actually does the job that you want. This is almost like talking to a fellow human, who can actually do all these capabilities for us, where our planners and schedulers get supercharged by focusing on what's most important: the decision of what content goes where.
Learn more: Intelligent scheduling with Amagi’s Smart Scheduler
How can content owners use agentic AI to reach new audiences?
Subramanian: We trained a machine to understand the art of microdrama — editorial capabilities, hooks, cuts. We took 30 minutes of content and were able to get 3-minute microdrama content created out of one single story. The system understands hooks, it understands what should be the editorial cuts, it understands what should be the 10 significant story elements it needs to stitch together to make that microdrama happen. The theme is: the richness that you have — how do we bring that richness to newer audiences, rather than just taking 16:9 television content and changing it?
The same thing happens in the news. Lots of news content comes in, but because of linear channel experiences, we are limiting the number of stories going to our platforms today. Both the speed at which you can do it, as well as the transformation of the stories to a Gen Z audience, are the future. As part of the first step, we launched Newspulse — Amagi's first agentic AI product. It takes all the broadcast assets that you have and converts them autonomously to TikTok, YouTube Shorts, Facebook, and everything you want to go to. You can talk like a producer to it — this is my audience, these are the sorts of stories I want to tell — and the system will start to understand the stories it has to pick, it understands the urgency, and is able to create those capabilities for you.
How does agentic AI handle data privacy for media operators?
Subramanian: There are two parts of the coin for AI — one is context, the other is privacy. Personalization and privacy go together. The more context about your organization, the better AI systems will do, not only understanding your terminology, your workflows, the specific aspects of how you see it, and your own capabilities. We are looking at that as a context that should be built. But more importantly, it's to protect that context in the environments of your own operating environment, so that it's not going out.
Privacy is at a role level that you can provide, as well as at an account level. With AI, it becomes more so, where your preferences, your understanding of the world that you live in, is captured inside your organization and not outside. When I talk about training on your content, it is also training on your intellectual property in terms of how you run your workflows. Everything is specific to a customer — we don't train on third-party data or third-party customers' content. That's not only for audio-video, but for all the data that you have as well. Privacy is an important part of the conversations that you need to have with any vendor, because the systems have to only react to your needs individually. That's super important, and I see that as part of the design process for every tool, every technology that we're building.
Want to go deeper on any of these topics?
German
French
Spanish
Korean
Japanese
Portuguese
