The local broadcast industry is undergoing a fundamental transition, shifting from a period of “playing with AI” to a rigorous era of operationalizing it. For the past year, the Local Media Association Broadcast Transformation AI Lab has sought to catalyze this shift.
Supported by the Google News Initiative, the lab brought together a cohort of 10 local broadcast companies — from large groups like Nexstar and Sinclair Media to independent players like Morgan Murphy and Capitol Media — to deepen understanding and drive implementation of AI systems and tools.
A key theme emerged: AI is not a replacement for local broadcast journalism, but it is a powerful tool to invert the cost structure of the business. This transformation is not merely about technological adoption; it is about institutional agility. Broadcasters who wait for “perfect” tools are effectively yielding the market to digital-native competitors. Success in this environment requires a “test-and-learn-fast” approach.
LMA has published a detailed AI playbook for local broadcasters, capturing the lessons and learning from the Broadcast Transformation AI Lab to help all broadcasters sharpen and accelerate their AI strategy. Here are six key opportunities identified by broadcasters through this collaboration.
1. AI for content versioning: The story as the atomic unit
The most profound realization emerging from the lab is the necessity of a radical economic pivot.
For decades, broadcasters have been weighed down by the capitalized costs of “feeding the beast” — the trucks, studios, cameras and crews devoted to the legacy news broadcast. It’s been an overwhelming challenge to reverse-engineer, reformat, re-version and distribute platform-optimized formats of stories built for traditional newscasts. In a multichannel content environment, where audiences want their news when and where they choose, these old workflows don’t work.
To survive, NOTA CEO Josh Brandau said that broadcasters must shift away from a newscast-centered model: “The story is the atomic unit of the business.” Everything else is infrastructure. AI facilitates the inversion of this model by automating the manual tasks that sit between the raw journalism and the multi-platform experience.
Examples of this shift abound. Cohort participant Pete Sockett at WRAL noted how his role has transitioned from traditional “television technology” to overseeing a unified stack of apps, web and mobile — an “everything” head of technology.
Nexstar Media Group’s implementation of NOTA offers one compelling proof of concept. They used the tool to bridge the gap between their broadcast ENPS scripts and WordPress-based digital sites, automating both the transcription and reformatting of TV scripts to web. Ron Parsons, Nexstar’s vice president of product solutions, estimated this workflow has saved between 15 and 30 minutes per story. Across 8,000 stories, he noted this projected to approximately 30,000 person-hours saved in a year. This is not just an efficiency gain; it is the equivalent of adding dozens of full-time reporters to the street without increasing headcount.
2. Beyond chatbots: Implementing goal-driven AI agents
Chatbots offer an easy and now-familiar entry point to leverage the utility of AI. But anyone who’s engaged in an endless series of AI prompts understands that chatbots also have their limits. The transformational opportunity for broadcasters is to implement AI agents that perform real work. These agents are workflow automators, not mere text generators. When used in enterprise accounts with established firewalls, they ensure that a station’s intellectual property is properly protected.
During the lab, several organizations shared examples of ways to use AI agents to automate key functions:
- Gray Media: Developed a “Sales Guidelines” bot that handles complex queries regarding advertising regulations, such as the varying state rules for cannabis or NFL broadcast restrictions. This allows account executives to get instant answers without interrupting mid-level managers.
- Morgan Murphy Media: Implemented a cross-platform proposal tool that analyzes audience data and matches it to specific client budget tiers, allowing for faster, more accurate sales pitches.
- KHQ/Cowles (Nonstop Local): Experimented with “NewsCoach” and “Stringer” agents to coach and critique story pitches based on the station’s specific style and ethics, and also to automate the creation of its hyperlocal newsletters.
These agents move AI from a novelty to a utility, ensuring the output remains consistent with the brand voice while protecting the newsroom’s internal data.
3. Supercharging sales productivity across the funnel
The adoption of AI in the sales department is an opportunity both for revenue preservation and margin expansion. Shannon Kinney of Dream Local presented a framework for using AI as a “junior sales assistant” — a tireless intern that handles the drudgery of research. Kinney noted she’s observed a 30% to 40% productivity lift among sales teams who smartly use AI.
A sales team’s “closing ratio” is a core KPI. Kinney enumerated the ways AI can supercharge sellers’ closing skills by refining each aspect of the selling funnel:
- CNA research: AI can instantly synthesize industry trends, such as interest rate impacts on mortgage brokers.
- Pitch critique: Sellers can use AI to “act as the boss” or “act as the client” to find holes in their proposals.
- Objection handling: AI can suggest closing questions that build trust and urgency based on specific client hesitations.
- Reporting narratives: Post-sale, AI can transform complex campaign data into “bite-sized” success stories for renewals.
4. Streamlining video production for FAST and social
Local broadcasters are under immense pressure to be multichannel, from live streaming and FAST channels to YouTube and online, as well as their traditional over-the-air distribution.
Etan Horowitz, Google’s global lead for AI for news, demonstrated how tools like Gemini and Google Vids can be used for “intelligent versioning” of broadcast video stories. A producer who might have spent hours clipping and posting newscast videos to alternative platforms in the past can now use AI tools, including ones from news-focused vendors like Genna and NOTA.
These tools enable a range of rules-based, AI-powered video clipping, formatting and distribution options that empower broadcasters to meet the “everywhere, all the time” demands of today’s audiences. A single producer can now manage multiple distribution channels effectively, competing with digital-native entities while maintaining the quality standards of a legacy newsroom.
5. Unlocking the ‘gold in the basement’: Archives and IP
Local broadcasters sit on decades of video archives that are largely unindexed and unmonetized, and in some cases degrading on old video formats and functionally unsearchable. Unlocking the value of this content has historically been prohibitively expensive, driven by the high cost of digitizing and cloud-storing video, as well as the cost and difficulty of tagging archived video to make it searchable.
AI tools change the business math around archives, creating new possibilities for both content creation and monetization. Costs of digitizing a station’s archives have decreased. In addition, AI’s ability to “understand” and semantically tag video content has advanced dramatically in the past few years. Unlike traditional keyword tagging, semantic-based metadata uses AI to understand the context and meaning of video (e.g., finding “all footage of 1990s hurricane damage” by recognizing visual patterns rather than just reading a file name). This transforms a basement full of tapes into a licensable, searchable revenue stream, in addition to preservation of historic archives.
WRAL has leveraged AI over the past several years in partnership with Eon Media to transform their archival video to enable new kinds of content and new revenue. It’s still early, but the station has already been able to more effectively search its own archives to create new original content, and WRAL is generating incremental revenue from licensing its archives.
Broadcasters who enable digitizing and semantic tagging of their archives also have another potential revenue stream — licensing these digital assets as training for AI companies. Erik Svilich, CEO of Encypher, cautioned that broadcasters needed to adopt IP protections and C2PA-compliant (Coalition for Content Provenance and Authenticity) defensive technologies like Encypher, Google SynthID and others to protect their content provenance.
These technologies use cryptographic metadata and digital watermarking to prove content origin. If a clip is scraped or misrepresented, this digital “DNA” provides the legal and technical proof of ownership required for licensing as well as for IP defense.
6. Trust as ‘moat’: Human-in-the-loop oversight
Trust is a local news broadcaster’s most valuable — and fragile — asset. As AI becomes ubiquitous, maintaining that trust is essential to maintaining a station’s viability.
A survey of more than 1,400 news consumers conducted by Local Media Association in collaboration with Trusting News found that 98.8% of audiences wanted a human to keep oversight over AI use for news.
While the survey found a number of AI use cases in newsrooms that had public support, the survey also identified areas where audiences are not yet comfortable with AI use in newsrooms. Notably, 30% of respondents expressed a baseline belief that AI should never be used in news.
Transparency around AI use is essential to maintain audience trust. Based on these research findings, Trusting News created an AI Trust Kit with audience-tested messages newsroom can use when AI is used in newsgathering or news reporting. Disclosure shouldn’t just state that AI was used, but why it was used and how it was verified. This “Human-in-the-loop” standard is non-negotiable for brand protection in an era where audiences regularly encounter automated “slop” and deepfakes.
The path forward: Institutional agility
The era of broadcast AI transformation is not about finding a single “magic” tool; it is about building an agile culture. The stations that lead the next generation of local media will be those that embrace structured experimentation and implementation, and place the story at the center of their business.
Local broadcasters possess two things AI-native competitors do not: trusted brands and deep hyper-local archives. By using AI to automate the infrastructure, leaders can clear the path for their journalists to do what they do best: tell the stories of their communities.
Editor’s note: AI tools were used to prepare this report, including to transcribe and summarize cohort sessions, and to draft an initial outline of key learnings.