Google’s AI chatbot ‘Gemini’ has received significant attention this year, topping its 'Year in Search' highlights. Here’s all you need to know about the new AI model, Gemini 3.
Google released its breakdown of the year’s top search trends, and topping the list – above Charlie Kirk, Pope Leo XIV, and Men’s Cricket scores – was its own Artificial Intelligence (AI) chatbot Gemini.
The company’s “Year in Search” highlights search queries that saw a significant and sustained spike in traffic in 2025 compared to 2024.
By those standards, and with a healthy dose of scepticism, the company’s top global trending search of the year was “Gemini,” followed by “India vs England” for men’s cricket, and “Charlie Kirk,” the US far-right influencer who was killed in September.
Google’s analytics show a spike in queries for Gemini starting in September, when the company unveiled a raft of AI updates across core services such as Chrome, Search, and Android.
Gemini’s upward trajectory
Google’s AI model has been the subject of mounting buzz this autumn, with the release of its latest iteration, Gemini 3 – powered by parent company Alphabet’s own proprietary chips.
It received high praise from OpenAI’s CEO, Sam Altman, leading global chipmaker Nvidia, and Salesforce’s CEO, who said he’s swearing off ChatGPT after trying Gemini.
“The leap is insane — reasoning, speed, images, video… everything is sharper and faster. It feels like the world just changed, again,” Marc Benioff wrote on X on November 23.
Among the new features, Google directly integrated Gemini 3 into its search engine through a native “AI mode,” and introduced “advanced agentic coding capabilities” and Google Antigravity, a new agentic development platform.
The company said Gemini 3 requires less prompting than previous models, and can “bring any idea to life with its state-of-the-art reasoning and multimodal capabilities.”
Most of the top independent benchmarks have confirmed Gemini 3’s reasoning power – Gemini 3 blew past OpenAI’s GPT-5 Pro to top the Humanity’s Last Exam benchmark, which measures general reasoning and expertise.
Its multimodal reasoning also scored the highest on benchmarks, including SimpleQA Verified, which measures progress on short-form factuality and parametric knowledge.
OpenAI declares a ‘Code Red’
Gemini 3’s rollout in November sent shockwaves through the Generative-AI ecosystem, putting pressure on industry leader OpenAI, which makes ChatGPT.
Already under fire from users over its latest model GPT-5’s “less friendly tone” and facing lawsuits from families who say ChatGPT contributed to loved ones’ suicides, OpenAI is now contending with increased competition from Google and Anthropic.
According to media reports citing an internal memo, Altman told OpenAI employees on December 1 that the company had declared a “code red” to improve the quality of ChatGPT and delay other products, including advertising, AI agents, and a personal assistant.
