Future of Marketing: Content Becomes More Valuable
Future of Marketing: Content Becomes More Valuable
A look at how AI generated content will impact enteprises.
This is the final article in our Future of Marketing series.
Image created on Midjourney.
Much has been said about AI lowering the quality of content on the Internet. That trend will continue, making original content much more valuable. In this new reality, content is unstructured data, and that data is gold.
Content has become valuable to those training their algorithms on original information. We see this with the increasing legal tension surrounding large-scale LLM training on copyrighted data. Before we dive into the legal implications, let us acknowledge that companies and organizations that create original data, research, and thought leadership content command more information value on the Internet.
Thanks to the rise of writing with LLMs like ChatGPT, Claude, Gemini, and Writer, content production and versioning are primary use cases, mostly for marketing but also across the enterprise. Original content fuels writing engines, honing their answers for industry, use, and brand-specific outputs.
This is as true of image and video generators as it is of their writing siblings. All these content generators have been trained on terabytes of public websites to learn how to create the best content that will meet human needs. All of them need better content to realize results befitting specific needs.
As a result of rising content value, enterprises seek to protect their unstructured data from AI exploitation and monetize their offerings. The legal enforcement of paywalls, monetization of original domain-specific data sets, and content creators suing AI Model developers for stealing intellectual property are all evidence of the rising value of original content in the AI world. Lawsuits against companies that own LLMs have demonstrated original content companies are not interested in allowing their intellectual property to be used as training fodder.
This has several implications; both for technology vendors who face changes in their business model, and companies that own and create original data in all its forms. See our CognitivePath Blaze on the legal implications created by AI training on original content here.
Original Content Gains Value
Image created on Midjourney.
Original content has become increasingly valuable as a source of data. [GV1] In addition, it should be noted in a world of robotic answers, original content stands apart from me, too homogenized with unique points of view information.
In a world where copycat and iterative content is accessible everywhere, it becomes regurgitated for all intents and purposes valueless. Paying customers and AI providers will seek out original data and thinking, including unique content. That value will come with a price. WIRED recently claimed, “Once controlled appropriately, data can yield the insights that make it more valuable than black gold.”
Why is original content so important? In the words of Pete Townsend on the Broken Record podcast, AI can only look backward. It cannot originate from a genuine new creation. AI must become sentient to create, and contrary to current hype from some vendors, most notably Open AI and X, most pragmatic visionaries don’t believe this will happen anytime soon.
In an ideal future state (cough), AI's best-generated content will be high quality, well-written, and thoroughly researched with original solid data. However, its output will be algorithmically designed to deliver the most probable result to meet the instructions used to prompt it. That means AI content is rooted in the past and anchored by this content without original thinking. The past is prologue with AI content, giving it a limited point of view. This creates a value ceiling for LLM-created materials.
This more significant macro value trend of AI-generated versus original content will exponentially increase the lower-end content creation we see. For example, bot-assisted content aggregation and rewriting will have a significantly lower value as time passes than original creative content and unique data. Unfortunately for human content creators, generative AI can create lesser quality content at a lower cost point, which will cause a significant minority of companies to simply increase their production as much as AI will allow them to scale.
Creating high-quality original content, from data sets and written content to visual information and video, will yield higher value. Unique points of view, new ideas, experience-based subject matter expertise, and, yes, real-world examples will all take on a higher value. But that’s not all, unique data extends to original imagery, videos, data sets, in world experiences and more.
One aspect of AI content generators is the expected reduction in workforce noted in the agency intelligence section. This reduction in human content creators – albeit the lowest performers – will put an additional premium price on high-quality original content.
A strong marketing strategy is needed to make content work regardless of format. As always, content is only valuable if it meets a need, and that requires brands to use content to serve a larger value proposition.
Investing in original content makes sense, depending on the marketing initiative. For some brand initiatives with an intelligent strategy in place, unique content will make winning campaigns for companies investing in original research, data, idea sharing, and solid brand journalism. But it must be of a quality level more significant than what AI alone can create. Depending on the size of the company and its appetite for hiring internal content creators, specialized agencies and freelancers that survive the impacts of AI on the sector may become ideal vendors to meet such needs.
For more day-to-day content needs like sales enablement, product support, go-to-market content, social media content, web pages, and communications, marketers may want an internal firewalled implementation or enterprise license of a content creation tool using original content and support data. When the messaging, brand guidelines, and strategy are in place, using human-piloted AI for derivative tactical content is more cost-effective.
Protecting Content As Data
Image by MysteryLab.
It costs money to hire people capable of generating original insights and information, and to build the very necessary governance processes needed to harness that data. Dr. Suzanne Weller, head of research at Privitar, said, “unstructured data will need the same degree of governance and privacy protection as has been achieved for structured and semi-structured data sources.”
In many cases, while guarantees are offered, cybersecurity risk might demand that you create a private instance to protect your data from getting ingested by a foundational AI model provider. Consider regulations in your sector and the potential risks of a customer data leak to make these determinations.
Then there is the business concern. Some organizations may want to protect their more public content from getting crawled to protect value. They may consider erecting paywalls and place no crawling code that bars LLMs from training on their websites to add a layer of protection.
In cases where the business's value proposition is directly tied to content, preventing AI and bot crawling, firewalling, and protecting content will become paramount. Expect many more gates and paywalls over the next three years. This is the hard work of implementing guardrails to protect your content assets BEFORE engaging in significant AI implementations.
There is an alternate scenario. Depending on what your organization tries to achieve with its content, there may be a monetization opportunity to make your data available to AI engines.
The Original Content Opportunity Summarized
As AI content generators like ChatGPT become more advanced, original high-quality content increases in value. AI systems need to be trained on large content datasets to function correctly. At the same time, the content they produce with their training data is always derivative and ultimately holds a lower value than the original new content.
Companies and creators behind this original content are starting to protect their intellectual property through paywalls, lawsuits against AI companies, and other means. Original, creative content based on unique data and insights will become more scarce and valuable. Brands, agencies, and creators that can continue generating high-quality original content in written, visual, and video formats stand to benefit financially.
Organizations that derive value from proprietary data must also consider new governance and security measures to protect their content as AI systems scour the internet for training data. Some may decide to allow access to shape how AI systems discuss their brand, while others will want to firewall and restrict access. Companies must weigh the costs, risks vs. rewards of original content strategies in an AI-powered world.
