Report: AI Pricing Over Time
The Dramatic Decline in AI Compute Costs & Strategic Implications for Enterprise
Executive Summary
OpenAI's recent 80% price reduction for its o3 API represents more than a competitive pricing move—it's the latest data point in an unprecedented 96% decline in state-of-the-art AI costs over just 27 months.
This analysis reveals a predictable technology deflation cycle occurring at 3-10% monthly rates, fundamentally reshaping the economics of AI adoption for enterprises. The implications extend far beyond cost savings to strategic questions about competitive advantage, innovation cycles, and business model adaptation.
(Mar 2023 - Jun 2025)
(Same Period)
(Consistently)
State-of-the-Art Model Cost Timeline
Methodology Note: This analysis tracks the cost of using the "best available" AI model at any point in time, showing how the frontier of AI capabilities becomes more affordable over time. Unlike tracking individual models, this shows the real cost evolution for organizations wanting cutting-edge performance.
| Time Period | SOTA Model | Input Price (per 1M tokens) | Output Price (per 1M tokens) | Cumulative Decline | Monthly Decline Rate |
|---|---|---|---|---|---|
| Mar 2023 - Nov 2023 | GPT-4 SOTA | $30.00 | $60.00 | — | — |
| Nov 2023 - Mar 2024 | GPT-4 Turbo SOTA | $10.00 | $30.00 | ↓ 67% / 50% | 10.4% / 8.0% |
| Mar 2024 - Jun 2025 | GPT-4o SOTA | $5.00 | $15.00 | ↓ 83% / 75% | 4.2% / 3.8% |
| Jun 2025 - Present | o3 SOTA | $2.00 | $8.00 | ↓ 93% / 87% | 3.5% / 3.2% |
Visual Analysis: SOTA Cost Evolution
State-of-the-Art AI Costs Over Time (Linear Scale)
Monthly Cost Decline Rates for SOTA Models
Strategic Implications for Enterprise AI Adoption
- Predictable Deflation: SOTA AI costs are declining at 3-10% per month, creating a predictable technology deflation cycle that enables strategic planning around continued cost reductions.
- Budget Planning Revolution: Traditional annual IT budgeting cycles cannot accommodate monthly cost reductions of this magnitude, requiring new financial planning approaches.
- Use Case Economics Transformation: Projects shelved for cost reasons 6-12 months ago should be immediately re-evaluated as economics have fundamentally shifted.
- Competitive Advantage Erosion: First-mover advantages in AI are shrinking rapidly as access costs plummet, shifting competitive focus to execution and integration speed.
- Infrastructure Strategy Shift: Building custom AI solutions becomes increasingly difficult to justify against rapidly cheapening state-of-the-art APIs.
- Innovation Acceleration: Lower costs enable more experimentation, creating a virtuous cycle where cheaper AI drives more innovation, which drives even cheaper AI.
Data Sources: OpenAI API Pricing, VentureBeat, India Today, Simon Willison's Weblog, Nebuly, and other industry sources
Report Generated: June 11, 2025 | Analysis Period: March 2023 - June 2025