I think this is the first time that ChatGPT has won me over with its features, rather than its performance, compared to Gemini.
Let me give you an example:
Today, Google released the exciting AP2 protocol. I first saw the news on several websites, but none of them cited the original blog post——and we all know that for every technology Google releases, they publish a corresponding introductory blog—— So, I used both ChatGPT and Gemini to search for the source of this information.
Here are the results:
ChatGPT:
Gemini:
Several things are immediately clear:
-
ChatGPT is more proactive than Gemini in using its web search function —— As far back as a year ago, I made multiple posts urging Gemini, a tool backed by the world’s largest search engine company, to deeply integrate with search to provide the best SEO experience. At the same time, I complained bitterly that even with the “Grounding with Google Search” feature enabled in AI Studio, Gemini was extremely lazy in calling upon it. But ChatGPT is different. It very quickly locked onto the news pages I had seen and found the original blog post with incredible speed and accuracy. Why can Gemini never seem to find it?
-
The precision of ChatGPT’s information and the timeliness of its knowledge base are things that Gemini currently cannot match at all——It’s as if ChatGPT is a living person, while Gemini is just a specimen. ChatGPT’s capabilities in image recognition, text generation, and programming might be on par with Gemini’s right now, or even weaker. But what good is that? ChatGPT is alive.
-
ChatGPT even offered multiple times to provide me with source links, download files, and GitHub repositories! —— I can hardly believe Gemini would do that for me!! Because Gemini simply can’t!!
-
Gemini may exhibit potential bias against Chinese——as shown in the image, when I ask this question in Chinese, Gemini does not perform a web search. Why is that?
Are these gaps that Gemini cannot close? At least in terms of the web search function, it’s entirely possible. When Gemini 2.5 Pro was released two months ago, the whole internet witnessed its power. I was so excited that I made a post praising Google. Google’s current product development capabilities are second to none, a leader in the industry, which shows they have the ability and the ambition to match any other company, and even surpass them.
However, since mid-August, Gemini’s performance has visibly declined. From Reddit to the official community forums, complaints about this decline are everywhere.
What on earth is wrong with Google?
Now, whenever I encounter a problem, especially one that requires searching or real-time data, the first thing I think of is opening ChatGPT, not Gemini. Two months ago, or even six months ago, I would never have done that.
As a Chinese American living in the United States, I frequently travel between China and the U.S. Since I still retain my Chinese language habits, I use a mix of Chinese and English in my daily conversations with AI. This discriminatory experience aligns perfectly with the current political climate, but it severely undermines technological equity and contradicts Google’s original technological mission.
