Islamabad, Mar 8, 2025: Google has upgraded its AI Overviews by integrating the powerful Google’s Gemini 2.0 AI Google’s Gemini 2.0 AI model in the US, promising quicker and more refined answers to intricate queries.
These improvements are particularly beneficial for coding, complex mathematics, and multimodal searches.
The company asserts that over a billion users are “engaging” with AI Overviews, largely due to its seamless integration into search rather than individual preference.
AI Overviews Now Mandatory for More Users
Previously, users could bypass AI Overviews by searching while logged out of their Google accounts.
However, the recent expansion now requires teenagers and unsigned users to engage with AI-generated summaries, effectively removing the opt-out option.
As a result, more individuals must now interact with these summaries, regardless of their preference.
AI Mode: A New Experiment in Search
Google is also rolling out AI Mode, an experimental feature accessible through Search Labs. AI Mode enhances AI Overviews by offering more advanced reasoning, detailed comparisons, and multimodal capabilities.
Positioned as the leftmost tab in the search interface, AI Mode allows users to pose intricate questions, refine responses, and access curated web links—at least for the time being.
This feature leverages a specialized version of Gemini 2.0, which Google promotes as an optimal tool for deeper research-based inquiries. 
Read More:
Realme 14 Pro Lite Offers Flagship Features at a Budget Price
However, since AI-generated results can sometimes reflect biases or subjective viewpoints, users may need to conduct multiple searches to confirm accuracy.
Upcoming Improvements and Testing Phase
Google’s Gemini 2.0 AI has plans to refine AI responses further by incorporating richer visuals, videos, and enhanced formatting.
Initially, AI Mode will be available exclusively to Google One AI Premium subscribers, with a wider rollout anticipated in the near future.
While AI-driven search enhancements are evolving rapidly, concerns about accuracy, reliability, and objectivity persist.
This makes extensive real-world testing an essential step before full-scale implementation.
