What Happened
OpenAI launched ChatGPT on November 30, 2022, as a free research preview. Built on GPT-3.5 and fine-tuned using Reinforcement Learning from Human Feedback (RLHF), ChatGPT was designed for multi-turn conversation and could answer questions, write code, draft essays, explain concepts, and engage in open-ended dialogue with a level of fluency that surprised even AI researchers.
Why It Matters
ChatGPT was the inflection point that brought AI into mainstream consciousness. It reached an estimated 100 million monthly active users within two months of launch — the fastest-growing consumer application in history at the time. It triggered:
- A global AI arms race among tech companies (Google, Meta, Anthropic, and others accelerated their AI efforts)
- Massive venture capital investment in AI startups
- Widespread adoption in education, business, and creative work
- Urgent policy debates about AI safety, regulation, and workforce displacement
- Microsoft's $10 billion investment in OpenAI
Technical Details
- Base model: GPT-3.5 (an evolution of GPT-3 with code training and instruction tuning)
- Training approach: Supervised fine-tuning on human-written dialogue, followed by RLHF (Reinforcement Learning from Human Feedback) using a reward model trained on human preference rankings
- Interface: Simple chat UI accessible via web browser
- Key innovations: RLHF made the model more helpful, harmless, and honest compared to raw GPT-3.5, while the conversational format made AI accessible to non-technical users for the first time
- Limitations: Knowledge cutoff (initially September 2021), tendency to hallucinate, difficulty with math and logic