Deal or No Deal?  License to Thrive in the new AI-Centric Media Landscape
LLM's want to make a deal for your content

Deal or No Deal? License to Thrive in the new AI-Centric Media Landscape

A significant AI driven transformation is already underway in the media sector:?

Recent OpenAI Licensing Deals

  • Associated Press: One of the first major deals, allowing OpenAI to use AP's extensive archives to train ChatGPT in exchange for licensing fees and access to OpenAI's technology.
  • Financial Times: OpenAI struck a deal with the FT to license its content for training models like GPT-4. The FT also became a customer of ChatGPT Enterprise for its employees.
  • Axel Springer: German media giant Axel Springer (owner of Politico, Insider) signed a deal with OpenAI, which some reports suggest could be a template for future agreements.
  • Stack Overflow: Partnership giving OpenAI access to Stack Overflow's database of over 59 million programming questions and answers to improve coding-related capabilities of GPT models.

Other Notable Deals

  • Apple: Reportedly offering at least $50 million over multiple years to partner with media companies and use their content for AI training.
  • Google: Demoed an AI tool called Genesis to major publishers like NYT and WSJ, aiming to generate news stories from facts.

This shift mirrors previous disruptive impacts seen in the entertainment industry similar to the rise of Netflix which put competitive pressure on the industry as it began to be the first place TV customers went to be entertained.?

AI is revolutionizing the media industry by being the new place customers go first for news and entertainment. The dynamic environment creates a complex array of challenges and opportunities for traditional media, prompting them to reassess the value and distribution of their content.

Signs that AI is owning the Media Customer

  • Over 60% of Gen Z and more than half of Millennials prefer AI tools for information gathering - The Verge
  • 54% of U.S. adults say they would be willing to get at least some of their news coverage from an AI system, with younger adults being more open to this than older adults - Pew Research Survey?
  • 32% of consumers are open to AI-written articles and 25% have already used AI tools to research news topics - Adobe Blog Post

What’s Underneath the Trend - AI's Competitive Edge in News Delivery

AI technologies offer unparalleled personalization and efficiency in news aggregation. AI Can:??

  • Personalize news delivery based on individual tastes and preferences
  • Quickly incorporate and synthesize information from multiple sources
  • Tailor news delivery to the specific task or context the user is working on

However, this innovation is not without its challenges.? AI can hallucinate eroding trust and it requires high-quality, human-generated inputs to produce high-quality outputs (for now).

"AI can generate news quickly but monitoring for accuracy is essential to maintain trust in journalism" - Columbia Journalism Review

Media Industry's Strategic Response

The media sector's reaction to AI's encroachment has been bifurcated:

Defensive:?

  • Publishers are closing down access to LLM’s web crawling and other methods of getting their content for free
  • Well-funded, high-quality publications like The New York Times want to fight in court to secure fair compensation for AI's use of their content.

Opportunistic:

  • Laggards and less traditional media companies strike licensing deals putting competitive pressure on industry leaders.

Lessons from Historical Media Transformations

Historical precedents, such as YouTube's creator fund and Netflix's content strategy, offer insights into managing the transition towards AI-centric media.

Netflix initially capitalized on low-cost, high-quality inputs from traditional media companies (e.g. Disney)? to secure their initial streaming customers. The company then leaned into its streaming-focused competitive advantage by personalizing content delivery and releasing entire seasons simultaneously—strategies traditional media could not match. Eventually, Netflix mitigated the backlash from their traditional media suppliers by producing its own content.

Similarly, YouTube's Creator Fund encouraged community growth and content creation which attracted more and more viewers.? Once YouTube owned the customer they began to harvest the ecosystem they started by changing the algorithm and pricing model with creators who couldn’t leave because YouTube had all of the eyeballs.

The Paradox of AI's Dependency and Media Companies Opportunity

Despite AI's advancements, its reliance on human-generated content presents a strategic opportunity for media companies to leverage their unique, high-quality content as a competitive advantage.

To thrive in this AI-driven era, media companies should focus on:

  • Owning the Audience: Build direct relationships with audiences through mechanisms like subscriptions and memberships to guard against being an commodity input for an LLM. Companies like the NYT will always be successful because they super serve their audience.? You should also look for the continuing rise of individual niche creators focused on Substack and similar platforms
  • Licensing High-Quality Content: Monetize proprietary content through strategic AI collaborations, ensuring a share in the AI-driven content ecosystem. Financial Times and StackOverflow offer early glimpses into Media companies catering to this new buyer
  • Differentiating Content: Invest in and highlight unique, expert-driven content that stands out in an increasingly AI-saturated market. The highest quality material that only humans are capable of creating will always be valuable

Christopher Wallace

MIT MBA | Regional Managing Director | Value Acceleration through Solving the "Last Mile" Problem in AI | I write about the measurable business impact of AI

6 个月

Wow so much to learn this space. Excited to see what happens next ??

Leonardo Coppola

Imprenditore SaaS e CEO @Voxloud | Aiuto le aziende ad automatizzare le vendite con l'AI in modo che possano crescere e scalare senza costi aggiuntivi | Ho fondato e scalato @Voxloud a 7 cifre partendo da zero

6 个月

data licensing deals is a great business model for AI

Really interesting read. Thanks for posting, Matthew! Do you think that licensing content that LLMs can learn on is the future, or do you see another way of getting models to learn gated content?

回复
Sebastian Illing

Fixing Email with GenAI ?? | CTO @ MailMaestro | Serial Entrepreneur ?? | Building with AI ?? for over 6 years

6 个月

Great insights, Matthew! Reddit also entered some data licensing deals lately.

回复

要查看或添加评论,请登录

社区洞察

其他会员也浏览了