Impact History

Explore the timeline of AI developments that shaped the world. View major milestones and breakthroughs in different formats.

View Mode

Model Release

GPT-5.2: OpenAI's Counterattack with Three-Model Architecture

OpenAI released GPT-5.2. Composed of three versions - Instant, Thinking, and Pro - it achieved human expert-level performance on 70.9% of professional tasks in the new 'GDPval' benchmark.

December 11, 2025→

Model Release

Claude Opus 4.5: Anthropic's Most Powerful Flagship Model

Anthropic released Claude Opus 4.5. Achieved 72.3% on SWE-Bench, claiming the top spot in coding performance. Extended thinking and hybrid caching deliver overwhelming performance on complex reasoning tasks.

November 20, 2025→

Investment

Microsoft Azure: The Only Cloud Offering Both Claude and GPT

Microsoft Azure becomes the only cloud platform providing access to both Anthropic's Claude and OpenAI's GPT models. Announced alongside a $45 billion partnership between Microsoft, NVIDIA, and Anthropic, positioning Azure uniquely in enterprise AI.

November 18, 2025→

Investment

AI Investment Boom: Largest Infrastructure Investment in History

Microsoft, Nvidia, Amazon, and Google announced over $400B total AI infrastructure investment in 2025. Unprecedented investment rush including massive Anthropic investment and Stargate Project for building AI era foundations.

November 18, 2025→

Model Release

Google Releases Gemini 3: Reclaiming Top Position in AI Benchmarks

On November 18, 2025, Google announced Gemini 3, surpassing Grok 4.1 to claim the top spot on LMArena benchmarks. Significantly improved multimodal capabilities and coding abilities, integrated into Google Search AI Mode from day one.

November 18, 2025→

Model Release

Gemini 2.5 Pro: Google's Serious Commitment

Announced Gemini 2.5 Pro at Google I/O 2025. Achieved performance to compete with OpenAI and Anthropic through 2 million token context, multimodal integration, and significantly enhanced agent capabilities. Deepened integration with Google Search.

May 21, 2025→

Model Release

Claude 4: Arrival of Next-Generation AI Assistant

Anthropic released Claude 4, significantly enhancing reasoning capabilities, context understanding, and multimodal processing. Further evolved the balance between safety and performance, establishing new standards for enterprise AI solutions.

May 15, 2025→

Model Release

Claude 3.7 Sonnet: The Pinnacle of Reasoning and Coding

Anthropic released Claude 3.7 Sonnet with Extended Thinking functionality. Achieved human-level performance on complex reasoning tasks and top-class scores on coding benchmarks.

February 25, 2025→

Model Release

DeepSeek-R1: AI Cost Revolution Shock

China's DeepSeek released reasoning-specialized model 'R1,' achieving performance comparable to OpenAI o1 at about 1/20th the cost. Historic release that overturned the 'scaling law myth' in AI development, opening a new era focused on efficiency. Shock waves through stock markets.

January 20, 2025→

Model Release

DeepSeek-V3: China's Open Source AI Shock

China's DeepSeek released the open-source model 'DeepSeek-V3.' Achieving GPT-4 level performance at low cost, heralding a cost revolution in AI development. Demonstrated Chinese AI companies' technical prowess to the world.

December 26, 2024→

Model Release

OpenAI o3: Quantum Leap in Reasoning AI

OpenAI announced the reasoning-specialized model 'o3,' demonstrating performance exceeding human level in advanced mathematics, coding, and scientific reasoning. Showed a new form of AI that visualizes thought processes and solves complex problems step by step.

December 20, 2024→

Model Release

Gemini 2.0: Dawn of the Agent AI Era

Google announced Gemini 2.0, taking a full-scale step toward practical AI agents. With multimodal generation (images, audio), tool use, and long-duration task execution capabilities, it heralded the arrival of an era where AI acts autonomously.

December 11, 2024→

Technical Breakthrough

Claude Computer Use: The Era of AI Operating PCs

Anthropic added Computer Use capability to Claude 3.5 Sonnet, enabling AI to operate computers like humans. Revolutionary feature controlling mouse, keyboard, and screen to automate complex workflows.

October 22, 2024→

Social Impact

AI Researchers Win Nobel Prize

Geoffrey Hinton and Demis Hassabis received Nobel Prizes, earning the highest honor in science for AI research. Hinton, the father of deep learning, won the Physics Prize, while Hassabis of AlphaFold won the Chemistry Prize, establishing AI's social standing.

October 8, 2024→

Model Release

Claude 3.5 Sonnet: Surpassing Opus Shock

Anthropic released Claude 3.5 Sonnet, achieving performance surpassing the top-tier Opus model with a mid-tier model. Dramatic improvements in coding, reasoning, and long-text understanding established a new standard for cost-performance.

June 20, 2024→

Model Release

GPT-4o: Realizing Omnimodal AI

OpenAI announced GPT-4o, achieving true multimodal AI that integrates text, audio, and images in real-time. A groundbreaking model that demonstrated the future of voice assistants and enabled natural dialogue with humans.

May 13, 2024→

Model Release

Meta LLaMA 3: Open Source AI Strikes Back

Meta released LLaMA 3 (70B and 8B models), demonstrating for the first time that open-source LLMs can match commercial models in performance. Accelerated AI democratization, providing worldwide researchers and developers access to state-of-the-art models.

April 18, 2024→

Model Release

Claude 3 Family Release

Anthropic released the Claude 3 family (Opus, Sonnet, Haiku). Claude 3 Opus surpassed GPT-4 on multiple benchmarks, establishing Anthropic as a true OpenAI rival. Achieved both AI safety and performance.

March 4, 2024→

Model Release

Sora: OpenAI's Video Generation AI

OpenAI announced 'Sora,' an AI that generates high-quality videos from text. Generating coherent videos up to 60 seconds long with astonishing quality that understands physical laws, it showed the potential to significantly change the future of video production.

February 15, 2024→

Model Release

Google Gemini Announcement

Google announced the multimodal AI 'Gemini,' clearly positioning itself as a competitor to OpenAI. With native multimodal design enabling integrated processing of text, images, audio, and video, AI competition entered a new phase.

December 6, 2023→

Social Impact

OpenAI CEO Dismissal Crisis

Sam Altman was suddenly dismissed as OpenAI CEO and returned just 5 days later. This incident highlighted the tension between 'safety vs. commercialization' in AI development and made AI company governance issues known worldwide as a historic event.

November 17, 2023→

Model Release

GPT-4 Turbo Release

OpenAI announced GPT-4 Turbo with a 128K token context window, cost reduction, and enhanced multimodal capabilities, significantly improving LLM practicality. This marked a turning point that accelerated the development of enterprise applications.

November 6, 2023→

Policy & Regulation

UK AI Safety Summit

28 countries signed the Bletchley Declaration, establishing the first international cooperation framework for AI safety. A historic international conference where major countries including the US, China, and EU participated to establish a collaborative system for AI risk assessment and mitigation.

November 1, 2023→