Studios and streamers have been under relentless financial pressure for years. Production costs keep climbing, subscriber growth has plateaued and traditional advertising revenues continue to contract. While the public debate remains fixated on generative AI creating synthetic video, the technology’s most immediate financial impact is happening behind the camera. Tools that once required massive post-production budgets are now accessible to virtually anyone.
Yet for major studios, unlocking that value has proven deceptively difficult. While point-solution AI tools can generate assets or speed up individual tasks in seconds, most entertainment organizations are hamstrung by fragmented systems, isolated data silos and manual handoffs accumulated over decades of mergers and acquisitions.
For the executives deciding where AI investment goes, capturing real ROI requires looking beyond isolated creative tools. The real opportunity lies in deploying AI agents that can orchestrate complex workflows across disparate systems, data, teams and external partners.
The legacy pipeline bottleneck
Most large media companies grew through successive waves of consolidation, accumulating decades of tech debt and fragmented sources of data that were never designed to speak to one another. Files sit in one asset management system, while rights and entitlements reside in another. Compliance review, viewership data, ad revenue and subscriber records are equally siloed across separate departments.
A question as straightforward as whether an archival catalog title can be monetized on an ad-supported tier in France often triggers weeks of cross-department email tag to verify content localization, check standards and practices, clear territory rights and confirm windowing restrictions.
Large parts of the media supply chain live outside the company walls, too. Localization, compliance, visual effects (VFX) and marketing are routinely outsourced to external partners, each running on their proprietary systems and isolated schedules. Consequently, a global release or licensing deal only moves as fast as a company’s external handoff.
Where entertainment leaders have bridged individual gaps with first-wave AI, the results have been immediate and substantial. Warner Bros. Discovery cut captioning times for unscripted streaming by up to 80% and slashed costs by up to half using generative AI tools. Netflix used generative AI across roughly 300 titles in the first half of 2026, primarily across post-production workflows.
These milestones prove that gen-AI delivers real bottom-line value. The next frontier is turning localized wins into compound gains: connecting rapid asset generation directly with rights clearances, compliance and international packaging so the entire supply chain moves together.
Enter agentic AI: orchestrating the supply chain
To understand where the technology is heading, it helps to distinguish between the first wave of generative AI and what comes next. While generative AI creates content and assets, agentic AI coordinates the systems, data and handoffs required to deliver them.
The path most entertainment organizations have taken so far relies on task-specific point tools: one system drafts subtitles, while another indexes scene metadata. But each operates in its own silo, outputting results to human coordinators who must manually transfer data into the next internal system.
Agentic AI workflows unify those fragmented steps. Instead of waiting for manual inputs at every stage, an AI agent is given a broader, human-defined objective — such as preparing an archival title for international syndication — and coordinates the necessary tools, databases and vendor handoffs required to execute it. Throughout the entire lifecycle, human leads remain central to creative judgment, editorial direction and legal sign-offs.
Consider localization, where a single global release moves across translation houses, standards review, legal clearances and technical mastering. Instead of coordinators manually chasing updates across disconnected teams, an agentic system manages the pipeline end-to-end — generating draft subtitles, validating territory restrictions and alerting human leads only when creative review or legal approval is needed.
The economic impact expands exponentially when applied to catalog monetization. Major studios sit on hundreds of thousands of hours of archival programming, much of it trapped in legacy vaults with incomplete metadata and ambiguous clearances. Agentic systems can ingest deep archives, enrich asset tags, match contract rights against real-time international FAST demand and package syndication-ready masters in hours rather than months. This is about removing the busywork so teams can focus on making great content and closing deals.
Connecting across the ecosystem
The true test for agentic systems happens when workflows extend beyond the company’s lot. Because entertainment relies on an extensive web of specialized external partners — localization, advertising and distribution — agents must be able to operate securely across corporate boundaries.
I’ve heard this many times from leaders across the media value chain, most recently in a meeting at CES. Marketing leaders from a major distributor and its ad agency intended to use AI to streamline audience measurement and hyper-personalized ad creation. But the initiative stalled because proprietary audience and asset data sat behind two strict corporate firewalls. Neither party could hand over raw database access. The only viable solution was enabling their respective AI agents to collaborate securely across both firewalls without exposing sensitive internal data.
Since media companies cannot, and should not, bring every specialized function in-house, the future belongs to an open agent ecosystem. Third-party vendors and agencies must be able to plug in as secure, modular extensions of a company’s internal pipeline, regardless of which underlying AI models or software stacks each party uses.
This cross-industry collaboration relies on emerging open standards rather than custom engineering. Standardized frameworks like the Model Context Protocol (MCP), which allows agents to securely read and interact with diverse enterprise data systems — alongside Agent-to-Agent (A2A) protocols — are laying the groundwork for a secure agent ecosystem. This allows independently built agents to communicate and delegate tasks to one another. Much like standard shipping containers unlocked global commerce, open protocols allow agents from studios, agencies and vendors to delegate tasks and exchange verified data seamlessly without putting proprietary IP at risk.
How media companies succeed
When media executives ask me about AI’s true return on investment, the conversation usually uncovers a fundamental deployment issue: an organization burning through an AI compute budget without moving the needle on production margins or release velocity hasn’t built an AI strategy, it has merely subsidized experimentation. Turning AI into tangible enterprise value requires the same operational discipline as any major infrastructure transition.
Technology alone won’t dissolve the invisible walls between departments that run on different schedules and priorities. Leaders need to bridge these gaps. That means establishing clear ownership for every end-to-end workflow, setting strict operational boundaries for what runs autonomously, defining how internal tools interface with external partners and building on open interoperability architectures.
We’re already seeing forward-looking broadcasters like Emtek Group embrace this blueprint. The company is connecting automated promo generation, live sports analysis and localization pipelines into a cohesive “Studio of the Future” strategy.
Commoditized AI creation tools will continue to proliferate, empowering everyone from seasoned showrunners to independent creators to boutique digital shops. However, traditional studios and established streamers possess the three assets that are hardest to replicate: massive content libraries, verified global rights portfolios and world-class creative talent.
By untangling legacy pipelines and letting agentic systems handle the operational heavy lifting, companies can eliminate administrative drag, protect production margins and unlock the full economic value of their content archives. This ultimately frees creative leaders to get back to what matters most: telling unforgettable stories.
Albert Lai is the Global Strategic Director for Media & Technology at Google Cloud.
