The clear difference between Gemini and AI mode will be an assistant. This is already the case today, and what the LLMs continue to make possible will only expand this gap. However, the similarities today ask me if this is confusing for most users.
9TO5GOOGLE has a REMOURCHED newsletter This highlights the biggest Google stories with additional comments and other songs. Register here!
I raised an eyebrow when Google AI mode last month began to allow subscribers to change models and Gemini 2.5 Pro. At the same time, Google published the “deep search” seven months after the Gemini application made the “deep search” available.
Earlier this week, AI mode began to test live search with a video entrance like Gemini Live, but without screen sharing. The last two features were pre-announcing at the 2025 E / S in May, but the AI mode adding “Canvas” surprised me a little.
Canvas is an interactive document that you can modify by highlighting a part and entering an prompt in the secondary cat. Gemini introduced this in March with an equal accent on text and coding.
The best answer I have obtained by speaking to Google recent launches is that people expect capacities such as in -depth research and the canvas from modern offers fueled by LLM. This makes sense, with the other new feature announced for AI mode being the possibility of downloading a PDF with your prompt.
In the near future, this duplication of capacities will continue with Project Mariner providing agent capacities in AI mode which can help you buy tickers, while it will be the “agent mode” for Gemini. In addition, the personal context will allow the AI mode to draw from your Gmail and other Google applications to take into account your preferences when response to prompts.
(By the way, the rate at which the new features of the AI mode are experimental are impressive, although Google desperately needs to expand world availability beyond three countries.)
Most people consider Google as a search engine, but it is ultimately a tool that must follow people’s times and expectations.
However, I am concerned about optics and I can absolutely see why people think that IA mode – and research to a large extent – should merge with Gemini to specify for people what is the AI application that they should use on Google. Gemini abandoning his distinctive mark and obtains a logo in four colors, certainly, did not help this.
Gemini does things that you would not want a tool as capable as Google Search to manage. On mobile, people know the idea of an intelligent assistant who can be used to control their phone and domestic devices, access their notes and files, check the calendars, make reminders, etc.
I think that separation is important because LLM assistants can do more for you, act being used to help in your daily life being the ultimate distinction.
It is still in a few years, and the best thing that the Gemini application can do so far is to publish these new generation assistant features as a more immersive phone check and a proactive experience. Frontier’s features like Deep Think yesterday are exciting for those who are on the edge of bleeding, but it is something like the example of Gemini E / S by seeing a test or another event on your calendar and helping you prepare that will really make the difference at home.
FTC: We use automatic income affiliation links. More.
