New paradigm?

[Sharing loose thoughts on the future of the websites, web traffic, LLM content, etc]

Bot traffic replacing human traffic: The internet is already crawling with many AI bots or agents scraping every corner of it because of the popularity of ChatGPT, Perplexity, Midjourney, etc., and AI co-pilots. Websites and publishers rely on human web traffic for monetization and engagement (views, ads, clicks, etc.). However, these AI bots or agents are increasingly developed to mimic human behavior. What will happen to websites that rely on human traffic when bot traffic replaces human traffic? Content creators will have to start monetizing bot traffic.

Higher value for new info: You can assume that most things on the internet that were freely available are already indexed and scraped, so the value of any new or gated information or news is high. Companies are racing to fence off their data (Reddit, etc).

The music industry has “back catalog” music and “new hits.” New information will be important to keep these models up to date. Perhaps “new hits” like breaking news will have a different rate card than “back catalog,” like historical information?

AI services will probably have to sign licensing deals to access this information. The result of NYT vs OpenAI/Microsoft lawsuit will impact this if/when it reaches the Supreme Court.

Licensing collective for independents: The worldwide web traffic follows the 80/20 rule - a small portion of websites account for most of the internet traffic. So, what happens to long-tail creators? Spotify essentially bootstrapped the popularity of Merlin, a collective for independent labels, for licensing simplicity. Warner, EMI, Sony, and Universal generally accounted for 80% of label share for most markets so could we see this emerge for longtail web publishers? A collective to negotiate rates with LLM bots or services.

Privacy vs Monetization: The launch of products like the Arc browser may muddle the conversation. They’re creating a browser just for you based on info/news from many websites without the user having to share their personal data. This is good for users, but bad for websites and publishers. This could kick off another round of pro-privacy-consumer vs. pro-monetization-publisher back and forths.

Others: It's easy to imagine how other areas can be impacted: the possibility of a new type of SEO for LLMs, monetizing page scrapes, redistribution of ad dollars, etc.

In my last post, I wrote about how we might need a new mobile architecture for future mobile-based consumption. Looks like we may have to reimagine the behind-the-scenes infrastructure for content consumption on the internet.