Statistically Biased Thoughts vol.1
By Mert Satilmaz
Although I hate the simplification of words by shortening them, will use AI for Artificial Intelligence in short for convenience.
Definitions
All artificial intelligence. Includes ML, computer vision, robotics, GenAI, agents, etc.
Subset of AI focused on content generation: LLMs, image, video, and code generation.
Some examples but not limited to: ChatGPT, Midjourney, Copilot, Claude, Suno, Sora
Subset of GenAI focused on autonomous action systems: AutoGPT-style agents, AI agents that execute tasks.
Some examples but not limited to: Clawdbot, OpenAI Operator, AutoGPT, Devin
AI Funding: 2000–2025
Now that we have the definitions and examples in place, let's look into how they have evolved historically and financially. All values are billions USD per year, not cumulative.
| Year | Traditional AI | GenAI | Agentic AI | Total AI |
|---|---|---|---|---|
| 2000 | 0.5 | — | — | 0.5 |
| 2001 | 0.6 | — | — | 0.6 |
| 2002 | 0.7 | — | — | 0.7 |
| 2003 | 0.9 | — | — | 0.9 |
| 2004 | 1.1 | — | — | 1.1 |
| 2005 | 1.4 | — | — | 1.4 |
| 2006 | 1.8 | — | — | 1.8 |
| 2007 | 2.2 | — | — | 2.2 |
| 2008 | 2.0 | — | — | 2.0 |
| 2009 | 1.7 | — | — | 1.7 |
| 2010 | 2.1 | — | — | 2.1 |
| 2011 | 2.8 | — | — | 2.8 |
| 2012 | 4.0 | — | — | 4.0 |
| 2013 | 5.5 | — | — | 5.5 |
| 2014 | 8.2 | — | — | 8.2 |
| 2015 | 12.1 | — | — | 12.1 |
| 2016 | 17.6 | — | — | 17.6 |
| 2017 | 23.3 | — | — | 23.3 |
| 2018 | 40.3 | 0.1 | — | 40.4 |
| 2019 | 36.5 | 0.2 | — | 36.7 |
| 2020 | 34.4 | 1.5 | 0.1 | 36.0 |
| 2021 | 65.3 | 2.5 | 0.2 | 68.0 |
| 2022 | 87.1 | 4.5 | 0.4 | 92.0 |
| 2023 | 89.0 | 29.0 | 2.0 | 120.0 |
| 2024 | 125.0 | 56.0 | 8.0 | 189.0 |
| 2025 | 116.0 | 87.0 | 22.0 | 225.0 |
Inflection Points
1997 — Deep Blue defeats Garry Kasparov
IBM Deep Blue defeated the world chess champion. This proved machines could outperform humans in narrow domains.
2012 — AlexNet wins ImageNet
AlexNet wins the ImageNet competition by a massive margin. This marks the birth of modern AI.
Growth in funding: $4.0B → $5.5B, 1.4× overall AI in one year.
Longer term impact: $4.0B → $40.3B, 10.1× overall AI between 2012 to 2018.
2017 — Transformer architecture invented
Google researchers published the paper “Attention Is All You Need”. This became the foundation for GPT, Claude, Gemini and all modern GenAI systems.
2018 — First commercially viable large language models
OpenAI released GPT-1. GenAI funding begins emerging from near zero.
2020 — Large scale models become commercially viable
OpenAI released GPT-3.
Growth in funding: $36.0B → $68.0B, 1.9× overall AI in one year.
2022 — The most important inflection point in AI history
ChatGPT launched publicly.
Overall AI funding: $92.0B → $120.0B, 1.3× growth in one year.
GenAI funding: $4.5B → $29.0B, 6.4× growth in one year.
2023 — GenAI becomes enterprise infrastructure
Microsoft Copilot, Google Gemini, and Claude enterprise deployments accelerate adoption.
Overall AI funding: $120.0B → $189.0B, 1.6× growth in one year.
GenAI funding: $29.0B → $56.0B, 1.9× growth.
Agentic AI funding: $2.0B → $8.0B, 4× growth.
2024 — Emergence of Agentic AI
Autonomous software agents such as Devin, autonomous research agents, and enterprise workflow agents begin deployment.
Overall AI funding: $189.0B → $225.0B, 1.2× growth in one year.
GenAI funding: $56.0B → $87.0B, 1.55× growth.
Agentic AI funding: $8.0B → $22.0B, 2.75× growth.
Unprecedented Growth Scale
With these being said, this represents unprecedented growth scale historically:
Grown 10.1× in 6 years between 2012 to 2018.
Grown 45× in 4 years between 2018 to 2022.
2022 → 2023 GenAI funding grew 6.4× in ONE year.
2022 → 2025 GenAI funding grew 19.3× in just 3 years.
Between 2023 to 2025 has grown 11× in 2 years.
Grown 56× in 13 years between 2012 to 2025.
Hopefully this helps everyone quantify the scale historically and financially.