Google Reveals New Deep Research AI Agent on Same Day OpenAI Unveils GPT-5.2

Overview

Both Google and OpenAI came up with major news on artificial intelligence on December 11, 2025. Google unveiled a new version of its Gemini Deep Research agent, whereas OpenAI revealed its new model, GPT-5.2. These changes demonstrate how fierce the competition has become in the field of artificial intelligence.

What Google Launched

Google’s new research agent is a redraft, more powerful version of Gemini Deep Research. It is based on Gemini 3 Pro, which is Google’s most advanced AI model capable of deep, complex reasoning tasks. This tool is not only aimed at report generation. The developers can be given access to the great research tools if the agent is embedded in their apps.

This

G

The instrument can perform information gathering on a massive scale; it can also handle an extremely intricate context and support work like testing drug toxicity or doing research for the right investment decision. At the heart of this ability is Google’s new Interactions API. This API enables third-party software to have deep interaction with the agentic AI.

Integration and Benchmarks

“After a while, Google expects to be able to integrate this new AI agent not only in its own services such as Search, Google Finance, the Gemini app, and NotebookLM but also third-party apps,” the article states. The company’s long-term goal is to build AI systems that can research and perform complex tasks beyond only searching.

Also, Google unveiled a brand new standard named DeepSearchQA, whose purpose was to put artificial intelligence to the test in multi-step, information-seeking challenges. On this standard, the new Deep Research agent was performing at a high level, although OpenAI’s latest model was still competitive for some tasks.

OpenAI’s GPT-5.2 Release

Simultaneously, OpenAI unveiled its latest large language model, GPT-5.2. The version is regarded by the company as the most powerful one with the features like general intelligence, coding capabilities, and comprehending long-context. GPT-5.2 has various levels, for example, Instant, Thinking, and Pro, to address the requirements of different users and developers.

GPT-5.2 is built to manage intricacies, for example, the invention of spreadsheets, attaining presentations, and dealing with multi-step projects. In the beginning, it will be accessible to those who have paid for the service, and developers through ChatGPT and the company’s API.

Competition and Timing

One of the most interesting things about the timing of these announcements is the fact that Google decided to make an announcement about its new research agent when OpenAI unveiled GPT-5.2. by the two companies are a race to the top in the field of AI should their advanced concentrated on cognition and integration and reach not only enterprise but also consumer users.

Major breakthroughs in AI research have been recognized through both initiatives, pointing out the directions tech giants intend to invest their efforts: Google braces for API integration and deep reasoning, while OpenAI aims at increased capacity and user-friendly improvements for professionals and developers. Competition between the two companies will most likely result in the acceleration of AI adoption in business and everyday tasks that will be the direct outcome of their ​‍​‌‍​‍‌​‍​‌‍​‍‌rivalry.