AI Revenue Growth 2026: Why 2 Days, Not 180, Is No Hype
In short
The AI industry now takes under 2 days to add the next $1 billion in cumulative revenue, down from roughly 180 days in 2023 - about 3 times faster than the internet or mobile eras. For Swiss companies, this means AI adoption delays no longer create linear disadvantages, but exponential ones.
The time the global AI industry needs to generate its next $1 billion in cumulative revenue has fallen from roughly 180 days in 2023 to under 2 days today. This figure covers revenue from apps, models and hosting, excluding China and chip manufacturers. That is a growth curve running roughly three times faster than the internet or mobile in their respective boom phases. Dismissing this as pure speculation misses the actual story: usage and willingness to pay are scaling for real - and that changes how much time Swiss companies have left to build their own AI roadmap.
180 days → <2 days
Time to add the next $1 billion in cumulative AI revenue, 2023 vs. today
What the 2-Day Figure Actually Measures
The acceleration refers to cumulative revenue from AI applications, model providers and hosting infrastructure - not valuations or funding promises. When the segment first crossed this billion-dollar pace in 2023, it took roughly half a year. Today it takes less than a weekend. For comparison: the internet and mobile eras were considered exceptionally fast in their own peak phases - the current AI curve runs roughly three times steeper still.
What the Major Market Analysts Independently Confirm
- Gartner (July 2026): The AI Platforms & Models market is forecast to grow 63 percent in 2026.
- IDC: Global AI spending is projected to rise from $235 billion (2024) to $632 billion (2028), a roughly 30 percent CAGR. Generative AI's share within that grows from 17.2 to 32 percent, reaching $202 billion by 2028 at a 59.2 percent CAGR.
- McKinsey estimates the annual value potential of generative AI across 63 analyzed use cases at $2.6 to $4.4 trillion.
- PwC projects an additional $15.7 trillion contribution to global GDP by 2030 - 14 percent above the baseline expected without AI.
Anthropic: From Startup to $30 Billion Business in roughly two years
The acceleration is most visible in a single provider. Anthropic's annualized revenue (ARR) stood at $87 million in January 2024. By December 2024 it had grown to $1 billion, by the end of 2025 to $9 billion - and by April 2026 to $30 billion, overtaking OpenAI's roughly $25 billion ARR. This is not a valuation story. It is paying usage multiplying within a handful of quarters.
AI bubble talk keeps circulating, but the revenue numbers are doing something previous tech waves never did this fast. Usage and willingness to pay are scaling in a way that looks more like infrastructure build-out than pure speculation.
Why This Isn't a Bubble - It's Infrastructure Build-Out
A bubble runs on expectation without payment. What's visible here is the opposite: compounding usage, compounding willingness to pay. That doesn't mean every valuation is justified or that every company survives - PwC's 2026 predictions explicitly call for precision over scattered initiatives and disciplined senior leadership. At the same time, entry costs for individual companies keep falling. Both trends - explosive growth at market level and falling costs at the adoption level - don't contradict each other. They reinforce one another.
What This Pace Means for Swiss Companies
For Swiss decision-makers, the relevant question isn't whether the market is growing, but how fast their own organization is keeping pace. Broader DACH adoption research shows a large share of companies remain at a novice stage of AI maturity even as active usage grows. When the global market moves in days rather than months, the gap between novices and mature adopters doesn't widen linearly - it compounds.
The Cost of Waiting Rises Daily
As long as the industry adds its next billion in under two days, a year of hesitation doesn't mean a year behind - it means a multiplying gap in maturity, data foundations and workforce capability.
Build-vs-Buy and Vendor Diversification as a Strategic Response
At this pace, building proprietary models is not a realistic option for most Swiss companies - and given the market shift from OpenAI to Anthropic within a few quarters, betting on a single vendor isn't a wise strategy either. A structured build-vs-buy decision with deliberate vendor diversification, as outlined in this analysis of AI strategy and global competition for Swiss SMEs, is the more resilient path.
Conclusion: Act, Don't Wait
The numbers refute the bubble narrative without ruling out caution. What they show clearly is that the fundamentals - usage, willingness to pay, market growth - are moving faster than most companies' internal planning cycles account for. Organizations that invest in a structured way now, rather than waiting, secure a lead that becomes nearly impossible to close in a market accelerating at this rate.
Frequently asked questions
- What does '2 days instead of 180' actually mean for AI revenue growth?
- The metric measures how long the global AI industry needs to generate its next $1 billion in cumulative revenue from apps, models and hosting (excluding China and chip manufacturers). In 2023 this took roughly 180 days; today it takes under 2 days - an acceleration running roughly three times faster than the internet or mobile eras.
- Is the current AI market a bubble?
- Available data points against a pure valuation bubble: growth is driven by paying usage, not just expectations. Anthropic, for instance, grew its annualized revenue from $87 million (January 2024) to $30 billion (April 2026). This doesn't rule out individual overvaluations or company failures.
- How fast is the global AI market growing according to analysts?
- Gartner forecasts 63 percent growth for the AI Platforms & Models market in 2026. IDC projects global AI spending rising from $235 billion (2024) to $632 billion (2028), a roughly 30 percent CAGR.
- What does this pace mean for Swiss companies?
- As the global market moves in days rather than months, the gap between companies that use AI in a structured way and the large share still at a novice stage widens exponentially rather than linearly. Waiting is not a neutral choice - it creates ongoing, compounding opportunity costs.
- Should Swiss SMEs build their own AI models or buy from vendors?
- For most Swiss SMEs, a build-vs-buy approach with vendor diversification is more sensible than in-house development. Given how quickly market leadership shifts - such as Anthropic overtaking OpenAI in ARR - deliberate vendor diversification reduces the risk of single-vendor dependency.
Sources
Would you like to explore this topic for your company?
Check Availability