OpenAI Speeds Up GPT-5.2 Launch as Google’s Gemini Gains Momentum

Affiliate disclosure: In full transparency – some of the links on our website are affiliate links, if you use them to make a purchase we will earn a commission at no additional cost for you (none whatsoever!).

OpenAI is preparing to release GPT-5.2 earlier than planned after Google’s new Gemini 3 model outperformed ChatGPT on major industry tests.

CEO Sam Altman called an internal “code red,” pushing teams to accelerate the launch.

The update may arrive as early as December 9, signalling one of the fastest competitive responses the AI industry has seen since ChatGPT first launched three years ago.

The urgency comes at a time when ChatGPT’s global user share is slipping.

Between August and November 2025, ChatGPT’s monthly active user share fell three percentage points to about 55%, while Google Gemini’s share rose by the same amount.

Although ChatGPT now has around 810 million monthly users, its growth slowed to only 6% in four months—an early sign of market saturation.

Altman is also scheduled to appear on NBC’s “The Tonight Show Starring Jimmy Fallon”, reflecting the high-pressure moment for OpenAI as competition heats up.


OpenAI Shifts Focus Back to Core Performance

 GPT-5.2

In early December, Altman sent a memo instructing teams to pause all non-essential projects. He asked everyone to focus only on improving ChatGPT’s speed, reliability, and reasoning.

Key changes under Altman’s “code red”

  • Work on advertising features paused
  • AI shopping agents delayed
  • Healthcare assistant projects halted
  • Personal assistant tool “Pulse” postponed
  • All teams reassigned to core ChatGPT improvements

Reports say GPT-5.2 already performs better than Gemini 3 in OpenAI’s internal tests. The update focuses on speed, stability, and custom options rather than new flashy features.

Also read about: OpenAI Updates Its ChatGPT Crawler ‘OAI-SearchBot’


AI Benchmark Race Tightens as Rivals Grow Stronger

Google’s Gemini 3 posted strong results on several tests. It scored 37.5% on Humanity’s Last Exam, far ahead of ChatGPT-5.1’s 26.5%.

It also set a new record of 1,501 on the LMArena leaderboard, where real users compare AI answers without knowing the model names.

Other competitors also add pressure. Anthropic’s Claude Opus 4.5 is getting attention from companies that want a more predictable and safe AI platform.

Tech leaders such as Salesforce CEO Marc Benioff even praised Gemini 3 publicly and said they were switching from ChatGPT.

ChatGPT still leads the market, with around 800 million weekly active users. But OpenAI remains unprofitable and now faces serious questions about its long-term strategy as Google, Anthropic, and other rivals move quickly.

Also Read:

On this page
Picture of Aishwar Babber

Aishwar Babber

Aishwar Babber is a dedicated blogger and skilled digital marketer with a deep passion for the latest tech trends and gadgets. His enthusiasm drives him to manage and grow TwinStrata, a platform dedicated to tech enthusiasts. With extensive expertise in digital marketing, SEO, and social media optimization (SMO), Aishwar thrives as a full-time marketer, successfully leading and contributing to various projects.