Cách mình dùng Google AI Mode và Gemini để tiết kiệm thời gian, rất dễ ai cũng làm được

Gemini 9 & Google AI Overview: A Detailed Usage Overview

Key Concepts:

• AI Overview (Google): A free Google Search feature providing summarized information with source links.
• Gemini (formerly Bard/Gemini 9): Google’s conversational AI model used for various tasks like event creation, information retrieval, and code assistance.
• Hallucination (AI): The tendency of AI models to generate factually incorrect or nonsensical information.
• Prompt Engineering: The art of crafting effective instructions for AI models to achieve desired results.
• API (Application Programming Interface): A set of rules and specifications that software programs can follow to communicate with each other.
• Tomica: A popular brand of die-cast toy cars.
• AntiGravity: A paid Google Workspace Labs feature utilizing Gemini for enhanced productivity.
• Google AI Ultra: The highest tier of Google AI subscription, offering access to advanced models and increased limits.

---

I. Introduction & AI Overview

The video details the speaker’s practical usage of Gemini (formerly known as Gemini 9) and Google’s AI Overview feature, emphasizing time-saving benefits compared to traditional methods. The speaker highlights the widespread integration of AI and aims to share effective techniques.

II. Utilizing Google’s AI Overview

The speaker begins by demonstrating Google’s AI Overview, a free feature accessible directly within search results. Instead of simply typing a query like “Snapdragon 8 Elite Gen 5,” the AI Overview provides a concise summary. Crucially, the speaker emphasizes using the “AI mode” for a more comprehensive and nuanced response.

Example: Searching for information on the Galaxy S26 reveals a summarized overview of leaked information, including a projected Unpack event date of February 25th, sourced from Vnet. This allows for immediate verification of information.

Key Point: The ability to trace information back to its source is a significant advantage, mitigating the risk of AI “hallucinations” (generating false information). The speaker stresses the importance of cross-referencing information, noting that rumors should be treated with skepticism.

Methodology: The speaker demonstrates iterative prompting – asking follow-up questions based on the initial AI response. For example, requesting a comparison of rumored S26 features with the Galaxy S25 in a table format. The AI successfully generates a comparative table.

Caution: The speaker points out that even AI-generated information can be inaccurate, citing a potential error regarding the Snapdragon 8 Elite 2 (believing it should be Snapdragon 8 Elite 5) and information sourced from Facebook.

III. Gemini for Information Gathering & Problem Solving

The speaker then transitions to showcasing Gemini’s capabilities.

Case Study 1: Finding Tomica Car Retailers in Singapore: The speaker recounts a challenge: locating stores selling Tomica die-cast cars in Singapore. Traditional Google searches yielded unsatisfactory results. Gemini, prompted with “Tomica vending machine location Singapore” (in Vietnamese), provided relevant locations. Specifically, it identified a location near Object Road, which the speaker confirmed had a vending machine.

Key Argument: Gemini excels at understanding nuanced queries that traditional search engines struggle with.

Case Study 2: Accessing Amazon App API Documentation: The speaker faced difficulty navigating Amazon’s developer documentation to find specific API information. Gemini, when provided with a description of the desired information, successfully located and provided relevant API links. The speaker confirmed the links were functional and useful.

Quote: “Thực ra là trước đó theo thói quen là mình đã tìm ở trên Google rồi. Tuy nhiên á các bạn biết sao không? Vấn đề là mình tìm mà nó không ra được cái nội dung mà mình mong muốn.” (Translation: “Actually, I had previously searched on Google. However, as you know, the problem was that I couldn't find the content I wanted.”)

IV. Gemini for Task Automation & Code Assistance

The speaker demonstrates Gemini’s ability to automate tasks.

Step-by-Step Process: Event Creation from Email: The speaker copied a list of events from an email and pasted it into Gemini with the prompt “tạo sự kiện lịch như sau” (translation: “create calendar events as follows”). Gemini automatically created corresponding events in Google Calendar, including details like dates, times, and locations.

Future Development: The speaker is developing a tool that will directly connect to a data warehouse, automating SQL query writing for reporting and data analysis. This tool is currently in testing and may be open-sourced in the future.

V. Gemini & AntiGravity: Advanced AI Tools

The speaker discusses their usage of AntiGravity, a paid Google Workspace Labs feature powered by Gemini. They utilize the Google AI Ultra subscription, which includes 30TB of Drive storage.

Comparison of AI Models: The speaker prefers Claude for code-related tasks, while Gemini is their primary choice within AntiGravity.

Benefits of Google AI Ultra: The speaker highlights the high limits associated with the Google AI Ultra subscription, eliminating concerns about usage restrictions.

Quote: “Hiện tại thì không có limit tuần nhưng mà ai biết có thể trong nó đổ lên ra.” (Translation: “Currently, there is no weekly limit, but who knows if it might be imposed in the future.”)

Workflow: The speaker uses the “planning mode” within AntiGravity for more detailed and controlled AI responses.

Cost-Benefit Analysis: The speaker believes the cost of the Google AI Ultra subscription is justified by the time savings and increased productivity it provides.

VI. Conclusion & Future Outlook

The speaker concludes by encouraging viewers to share their experiences with different AI tools like Claude and Gemini. They emphasize the importance of verifying AI-generated information and highlight the potential of AI to streamline workflows and enhance productivity. The speaker plans to share further insights on AntiGravity in future videos. The overall takeaway is that Gemini and Google’s AI tools are powerful resources for information gathering, task automation, and problem-solving, but require critical evaluation and thoughtful prompting.