Hi there,
If you work in digital media, a big part of the job today is to consider the ways that generative AI could impact your business. How it might, for example, enable more efficient research or content creation – but also how it’s threatening to disrupt platform distribution and certain publisher business models.
If you run a generative AI media startup, though, the calculus is different. The main job is to imagine a future unconstrained by existing models, a future where upending the way things work today is the point.
In an interview last week, Adam Mosam, cofounder of the generative AI video startup Channel 1 News, painted one such vision:
Imagine watching CNBC, except what you’re looking at is analysis of stocks that are in your portfolio, or industries that you’re already watching, or if you’re watching sports, it can go more in depth on the teams that you love, as opposed to waiting for the parts of the content that you’re really interested in.
Channel 1 News plans to launch later this year with a 30-minute weekly news show distributed through a FAST channel. Ultimately, the company’s ambition is to use AI to create personalized newscasts, like the one described above.
It’s an intriguing vision, but there are, of course, challenges:
Unique content & data: As David wrote last week, publishers with unique content & data are well-positioned to create valuable products on top of open source AI models. But Channel 1 lacks original content and plans to rely on press releases and partner content as inputs to generate its scripts. While this is a reasonable approach to start, both of these sources come with challenges: Press releases are dry and not journalistic, and other news outlets may be hesitant to license their content.
AI-native formats: Channel 1 cofounder Scott Zabielski admits that their AI-generated anchors aren’t yet realistic: “The elephant in the room is this still looks like you’re watching a video game character talk.” But even as the technology improves, it’s unclear whether a personalized, anchor-based newscast will make sense. In fact, it’s more common that, as a platform develops, new native formats take hold and replace existing ones.
Video distribution: Launching on a FAST channel will help Channel 1 reach an audience quickly, but the company’s long-term goal is to create its own platform. While it’s an admirable goal, few media organizations have succeeded here, with most relying on existing video platforms for distribution.
How Channel 1 navigates these questions will, in large part, determine its success. But, at a high-level, its thesis that AI-enabled personalization will impact news creation & consumption is smart. When it launches, the company will be one among many experiments testing what this thesis looks like in reality.
And here’s the latest news in digital media:
Publishers:
What’s New in Publishing screenshotted 36 publisher paywalls.
The AP can reset its OpenAI deal if another publisher secures more favorable terms.
Maria Ressa is leading an international committee to create a charter on AI usage in media.
WaPo released its 2023 workforce demographics report.
Adweek appointed Will Lee as its new CEO.
Platforms:
Artifact is adding an AI-powered, text-to-speech feature.
New research paints a complicated picture of how Facebook shapes political beliefs.
The NYT analyzed how 15 celebrities are using Threads.
Reddit is testing a new “Official” label on verified accounts.
Twitter (now, X) officially rolled out its creator rev share program. The platform also reinstated Kanye West’s account.
Entertainment:
The Emmys will be postponed this year because of the actors’ and writers’ strikes.
Hollywood studios are hiring for more AI jobs, including a $900K AI product manager role at Netflix.
ESPN is testing an all-female SportsCenter.
Thanks for being a part of Business Side’s public beta. Have a great day!
Was this forwarded to you? Subscribe here.