AI Search: How AI Risks Undermining the Free Internet
Written on
The Impact of AI on Internet Usage
As I delve into research these days, I've noticed a significant decline in my visits to websites. The reason? Like many others, I now turn to ChatGPT 4 for quick summaries. What used to take me an hour can now be condensed into a mere five minutes. With the help of a plugin, it can even access the web in real-time, beyond the limitations of pre-2021 data.
When I need credible information, I ask ChatGPT to reference sources, and I might explore those sites if something piques my interest—though that happens only about 20% of the time.
Consider the ramifications of this shift on the internet's revenue model. If users are less inclined to browse websites, who will click on advertisements? This trend poses a threat to content creators, as dwindling revenue could lead to fewer updates or even the disappearance of some sites.
Ironically, AI search relies on the very content it risks jeopardizing. Luckily, there are potential solutions, but content creators will need to adapt, and AI platforms must also play a role.
To illustrate my point, I’m including a delightful chocolate fudge cake recipe—because why not?
Examining the Future of Search
Is this truly the future of searching for information? Let’s conduct a little experiment. If I were to search for an excellent chocolate fudge cake recipe, here’s what I might find from Google:
(Screenshot of a Google Search for Cake Recipe - Screenshot by author)
In comparison, here's a straightforward recipe provided by ChatGPT 4:
Chocolate Fudge Cake Ingredients:
- For the Cake:
- 2 cups all-purpose flour
- 2 cups sugar
- 3/4 cup unsweetened cocoa powder
- 2 teaspoons baking powder
- 1 1/2 teaspoons baking soda
- 1 teaspoon salt
- 1 cup whole milk
- 1/2 cup vegetable oil
- 2 large eggs
- 2 teaspoons pure vanilla extract
- 1 cup boiling water
- For the Fudge Frosting:
- 1 cup butter, softened
- 1 1/2 cups unsweetened cocoa powder
- 5 cups confectioners' sugar
- 1/2 cup whole milk
- 2 teaspoons pure vanilla extract
- 1/2 cup hot water
Instructions:
- Preheat your oven to 350°F (175°C). Prepare two 9-inch cake pans by greasing and lightly flouring them.
- In a large mixing bowl, blend the flour, sugar, cocoa, baking powder, baking soda, and salt.
- Incorporate the milk, vegetable oil, eggs, and vanilla into the dry ingredients, mixing until smooth.
- Gradually mix in the boiling water. The batter will be thin, which is perfectly fine.
- Distribute the batter evenly between the prepared pans.
- Bake for 30 to 35 minutes, or until a toothpick inserted in the center comes out clean.
- Allow the cakes to cool in the pans for about 10 minutes before transferring them to a wire rack to cool completely.
While the cakes cool, prepare the fudge frosting:
- Beat the softened butter until creamy.
- Gradually mix in the cocoa powder, confectioners' sugar, milk, and vanilla, blending on low speed until combined.
- Slowly add hot water, one tablespoon at a time, until the frosting reaches your desired consistency.
Finally, frost the cooled cakes. You can layer frosting between the cakes and on the top and sides. Enjoy your delicious chocolate fudge cake!
Is this the ultimate chocolate fudge cake recipe? Perhaps not, but it’s likely delightful and certainly adequate.
The Benefits of AI Search
One major advantage of AI search over traditional Google searches is the ability to ask follow-up questions about the results, making it feel more like a research assistant than just a search engine.
Google is aware of this shift in user behavior. They are diligently developing AI-enhanced search capabilities to address these concerns.
> "With new generative AI capabilities in Search, we're now taking more of the work out of searching, so you'll be able to understand a topic faster, uncover new viewpoints and insights, and get things done more easily."
> — Supercharging Search with Generative AI, Google
The Subscription Model's Viability
Not all content creators rely on advertisements; many writers on platforms like Medium prefer paywalls. However, integrating AI search with subscription models will require effort. Content must be accessible to search engines to ensure visibility.
What’s to prevent this content from being included in AI-generated summaries? While some users may prefer the original content, many will not. Medium, for example, allows users to read three articles for free, which might be sufficient for casual browsers.
One potential solution is for AI companies to license content from providers and track page access. Revenue from licensing could then be shared with content creators. However, this would necessitate forming commercial partnerships with leading AI search providers—something only larger sites like Medium might find feasible.
Challenges for Ad-Supported Content
What about creators who depend on ad revenue? If an AI search engine summarizes content, it must also cite sources and display advertisements. This could be straightforward if the AI picks up ad links from original content.
However, challenges may arise regarding the prominence of ads and the preferences of advertisers. Additionally, content providers would likely need to agree to the terms set by AI search engines.
A significant advantage here is that this approach modifies an existing business model rather than creating a new one.
How Affiliate Marketing Can Thrive
AI summaries could also incorporate affiliate links, which could be simpler to manage than ad placements. Unlike ads, no formal relationship is necessary between the content provider and the AI platform—cooperation is sufficient.
What Lies Ahead
There may be a niche market for licensed access to sites like Medium, but it would be limited. Only content with substantial value would attract the interest of AI search platforms.
Incorporating ads and affiliate links appears to be the most straightforward approach. AI companies must ensure fairness in presenting these ads within the search summaries.
Perhaps it’s time to establish a standard for sharing this information—if one doesn’t already exist.
The Future of SEO
If AI search engines begin integrating ads and affiliate links into their summaries, the landscape of SEO will undoubtedly become more complex. It will no longer be enough to optimize for Google or Bing through keyword selection alone. Content must also align with how large language models evaluate queries, a process that may remain opaque even to AI platform developers.
In this video, former Google executive Mo Gawdat discusses the potential dangers of AI and its implications for our future. Don't miss out on his insights!
This video explores how AI is transforming the internet and the potential consequences of this shift. A must-watch for anyone interested in technology and its future!