Content
summary Summary

For several months, Google has been testing the Search Generative Experience (SGE), a new form of Google search that combines AI-generated content with traditional link search. Now, a new version has surfaced.

For Google, AI search offers a chance to fend off new attackers in the search business, such as Microsoft and OpenAI. It could also transform Google from an open web platform to a closed content platform, similar to social media platforms, but with a focus on AI content. This is where the long-term revenue potential lies.

But Google also faces the challenge of balancing the high cost of training, deploying, and scaling AI models with the revenue potential of AI search. Standard Google search is a very profitable business, with high margins per average query and billions of queries per day.

Converting this to new AI search methods without loss and even with the potential for growth could be an insurmountable challenge, at least in the short to medium term. With the Search Generative Experience, Google is aiming for a smooth transition between two business models.

Ad
Ad

The pressure is off Google

So Google may be trying to be a little faster than the competition, but only as fast as it needs to be to avoid falling behind in a potentially emerging online ecosystem centered around chatbots.

This thesis is supported by a new variant of the "Search Generative Experience" (SGE), which SEO specialist Glenn Gabe posted on Twitter and calls "SGE lite": It appears much more stripped-down than previous AI results.

The full AI answer only appears after a second interaction with a "show more" button. The classic links are much higher up - and with them, Google's ads could slide back to the top, with lower costs for AI generation at the same time.

The "SGE Lite." | Image: Glenn Gabe via X
The SGE shown so far displayed significantly more AI content on monitors, pushing traditional web links and ads to the bottom. | Image: Google

Microsoft's experiment with Bing Chat has shown that a chat offering by itself does not lead to growth in the search market. Microsoft CEO Satya Nadella recently admitted that he overhyped Bing Chat.

OpenAI's browsing function for ChatGPT still lags classic search engines in terms of reliability and functionality, while at the same time incurring high costs for the provider.

Recommendation

So the pressure on Google to innovate, created by ChatGPT's massive success and Microsoft's offensive earlier this year, might be easing.

Ad
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.
Ad
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.
Support our independent, free-access reporting. Any contribution helps and secures our future. Support now:
Bank transfer
Summary
  • Google has been testing the Generative Search Experience, a new form of Google search that combines AI-generated content with traditional link search, for several months. Now, a new variation has gone live.
  • With generative AI search, Google can fend off new competitors in the search business, such as Microsoft and OpenAI. But Google could also evolve from an open web platform to a closed content platform, similar to social media platforms, but with a focus on AI content and long-term revenue potential.
  • The challenge for Google is to balance the enormous costs of training AI models and operating costs with potential revenues that must be at least equal to, if not greater than, current revenues.
Online journalist Matthias is the co-founder and publisher of THE DECODER. He believes that artificial intelligence will fundamentally change the relationship between humans and computers.
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.