Reed McGinley-Stempel is the co-founder and CEO of Stytch.
As artificial intelligence (AI) continues to take off, the landscape of data utilization is undergoing a radical transformation, creating new opportunities (and fears) for the data providers that many large language models (LLMs) have been historically trained on. Specifically, AI’s attribution models (or lack thereof) threaten the fundamental underpinnings of online advertising. Companies dependent on attribution-based ad revenue need to start acting now to protect themselves.
In this article, I’ll go over how AI is disrupting typical attribution models in online marketing and how the companies dependent on them can take action before it’s too late.
AI And Attribution
Throughout most of the 2000s, digital platforms like Yelp, Quora, Expedia, Reddit and countless others depended heavily on an ad-driven revenue model. Visitors to these sites were exposed to a slew of targeted advertisements, their clicks generating invaluable revenue streams for these platforms. However, AI’s meteoric rise has disrupted this tried-and-true monetization model, largely due to the shift toward a first-party captive experience facilitated by AI-powered services like ChatGPT.
Consider a scenario in which you ask ChatGPT for the best place to get a steak in New York City. By default, the AI will generate a list of top-ranking steakhouses, devoid of any references to the data sources from which it drew this information. The reality is that the rankings probably come from Yelp or similar platforms—a fact lost in this interaction.
That lost fact translates to lost attribution and lost money for Yelp.
By bypassing the need to visit the original data source (in this case, Yelp), ChatGPT eliminates opportunities for ad click-based monetization and affiliate link revenues. More pertinently, if there’s no paid data aggregation agreement in place, this interaction diverts revenue away from Yelp’s employees and shareholders. As intelligent services like ChatGPT continue to improve, this will likely divert significant amounts of traffic away from aggregation- and attribution-dependent companies.
This stark difference in experience is more evident when compared to the pre-AI era of search engines. Back then, a query like this would have led users directly to the data source’s website, ensuring that these platforms could monetize their ad clicks and affiliate links. However, AI’s immersive first-party captive experience has disrupted this revenue stream.
At the same time, companies that have historically made their data free to the public are realizing its monetary potential. Amid fierce backlash, Reddit CEO Steve Huffman announced the company would begin charging for access to their API—setting off a maelstrom of protests, subreddits going dark and several companies announcing their inability to continue contracting with the company.
Whether or not you agree that Huffman’s decisions were justified, the fact that a company like Reddit—with a huge fanbase and a community formerly built on the ideals of open source and open access—is making such drastic changes to its business models indicates how sizable AI’s threat and monetary potential are.
So, what’s to be done?
Ways To Take Action
Fortunately, for most technology leaders, there are more moderate approaches to protecting your data and open endpoints. Rather than seeing AI as somehow adversarial to data attribution revenue models, we believe this step change in AI capabilities is more of a wake-up call for companies to arm themselves with the right technical tools and protocols to maintain control over their data assets and preserve their business models.
Today, there are a number of measures and tools in fraud, risk and access control that can help. Here are a few:
1. Device Fingerprinting: This helps identify traffic and tier access based on a variety of cues.
2. CAPTCHA: These tools can be effective at stopping bot activity, but be aware that basic CAPTCHA solutions are now less effective in the face of more sophisticated AI.
3. Additional Protection Layers: Some companies offer additional protection layers that help stop high-velocity attacks, but it’s worth noting these solutions can’t capture other data-scraping methods like headless browsing.
Conclusion
Instead of looking at this situation as a war, it’s perhaps more productive to view it as a wake-up call for conventional internet companies to evolve their data management strategies. The AI boon doesn’t have to mean revenue doom. With the right controls in place, it’s entirely feasible for companies to enjoy the benefits of AI advancements while also preserving—and potentially enhancing—their existing monetization models.
As we navigate through the AI era, this constructive approach toward adopting stringent control measures will not only help mitigate the growing tension between data owners and AI platforms but also pave the way for a more balanced, efficient and equitable data ecosystem.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Read the full article here
