Pick your fighter — Google or ChatGPT: A new experiment pitting Google’s and ChatGPT’s search functions against each other across different categories of search requests has produced an unexpected outcome — calling into question where the AI industry’s crown jewel stands in its race to dethrone the longtime king of search, Google.

The breakdown: The experiment saw 62 queries tested on both Google’s and ChatGPT’s search engines, with performance compared across a number categories like informational, commercial, comparison, disambiguation, and medical searches. Results were evaluated based on the degree to which answers matched up with user intent, alongside errors, omissions, and weaknesses in the answers.

In a somewhat unexpected turn of events, Google came out on top. The old-school search engine particularly excelled in informational, navigational, local, and commercial searches, with the platform’s wide index of web pages, longtime trove of data on user intent, and supporting tools — like Google Maps for local queries — providing a superior search experience for users across the (albeit small) sample of queries used in the experiment.

But ChatGPT did fight back. OpenAI’s platform stood out in content gap analysis, which identifies areas of improvement in a piece of content — a crucial task for content creators and SEO professionals. Google’s standard search functionality didn’t stack up in this area, since it only shared search results related to the topic rather than filling in the gaps.

There was definitely room for improvement, though. ChatGPT exhibited some trouble with accuracy and omissions in its responses — highlighting a need for improvement in fact-checking and information retrieval. If you thought the AI platform was the all-knowing artificial genius, this experiment suggests you might want to reevaluate.

Tags: