ChatGPT Hit 100M Users in 2 Months. Here's What That Means for Venture.

The telephone took 75 years to reach 100 million users. Instagram took 2.5 years. ChatGPT took 2 months. OpenClaw — an open-source AI agent framework launched in November 2025 — hit 100,000 GitHub stars in 48 hours and 250,000 stars within four months, surpassing React as the most-starred software project on GitHub.

These are not just interesting data points. They are signals about the structural change in how technology value is created and captured — and that change is now colliding directly with venture capital.

NUVC is an AI-native venture intelligence platform built by two founders using AI tools (no engineering team). We track these adoption curves because they directly determine which companies are worth backing, at what stage, and with what thesis. Here is the data and what it means.

How Does AI Adoption Speed Compare to Social Media?

Comparing time-to-100M users across platform generations reveals three distinct adoption paradigms:

Infrastructure era (1870s-1990s):

• Telephone: 75 years (1876)

Social platform era (2003-2023):

• LinkedIn: 5 years to 100M (2003 launch) — now 1B+ members, 14% year-on-year growth
• Facebook: 4.5 years to 100M (2004 launch) — now 3.07B MAU
• Instagram: 2.5 years to 100M (2010 launch) — now 3B MAU (2025)
• TikTok: 9 months to 100M (2017 launch) — now 1.92B MAU
• Xiaohongshu (RED): approximately 6 years to 200M MAU (2013 launch) — now 350M MAU, targeting 700M
• Threads: 5 days to 100M (July 2023)

AI intelligence era (2022-2026):

• ChatGPT: 2 months to 100M (November 2022 launch) — now 800M weekly users, $25B ARR
• DeepSeek: approximately 2 weeks to 100M users in China (January 2025) — now 97M MAU globally
• OpenClaw: 48 hours to 100K GitHub stars (November 2025) — now 250K+ stars, most-starred on GitHub
• Claude Code: $0 to $1B ARR in 6 months (May 2025 launch) — now $2.5B ARR, most-used AI coding tool
• Cursor: $0 to $2B ARR in approximately 2 years (2023 launch) — 60% enterprise penetration

The trend is unmistakable. Each generation of technology reaches scale faster than the previous one. But the reason for each acceleration is different — and that matters enormously for how you evaluate companies.

What Is a Capability Effect vs a Network Effect?

Social platforms grew through network effects: each additional user made the product more valuable for existing users. More friends on Facebook meant more reasons to check Facebook. The growth curve followed a classic S-curve — slow start as the network built, rapid inflection once critical mass was reached, plateau as the addressable market saturated.

AI tools grow through capability effects: each improvement to the underlying model makes the product better for all users simultaneously, regardless of how many users there are. When Anthropic improved Claude's reasoning capabilities, every Claude user got a better product overnight. There is no network to build — there is only the model to improve.

This creates a fundamentally different growth dynamic:

• Social platforms: Slow start, fast middle, retention via social graph lock-in. Switching cost is losing your connections.
• AI intelligence tools: Instant adoption because the capability is immediately useful to individuals. Retention depends on improvement speed. Switching cost is learning curve and workflow integration.
• AI developer tools (Claude Code, Cursor, OpenClaw, v0): Adoption tracks how fast developers switch their daily workflow. This is the fastest B2B adoption category in history — Claude Code reached $0 to $1B ARR in 6 months.

What Is the Current Scale of the Leading AI Platforms?

As of March 2026, the AI intelligence platform market has produced growth rates that have no historical precedent:

• OpenAI (ChatGPT): 800M weekly users. $25B ARR. Grew from $2B to $20B ARR in 12 months. For context, Google took 8 years to reach $20B annual revenue.
• Anthropic (Claude): $19B ARR — up from $1B just 14 months ago. $380B valuation. Projects $70B revenue by 2028.
• Claude Code specifically: $2.5B ARR. 46% developer preference in recent surveys, surpassing GitHub Copilot and Cursor as the most-used AI coding tool.
• Cursor: $2B ARR (doubled in 90 days). But active switching to Claude Code is accelerating. 60% enterprise penetration shows the B2B path.
• OpenClaw: 250K+ GitHub stars in 4 months. Most-starred software project on GitHub. Nvidia described it as "to agentic AI what GPT was to chattybots."
• DeepSeek: 97M MAU. Removed financial and technical barriers to advanced AI in underserved markets — the first AI platform to demonstrate that capability effects can scale outside the US market.
• GitHub Copilot: 4.7M paid subscribers, 90% of Fortune 100 adoption. But market share is stagnant against the wave of newer tools.

95% of developers now use AI tools at least weekly. 75% use AI for more than half their coding tasks (Pragmatic Engineer, 2025 developer survey).

How Does Faster Technology Adoption Change What Venture Capital Funds?

Five structural changes are underway:

1. Build speed is 10x faster. Vibe coding with Claude Code, Cursor, OpenClaw, and v0 means a technical founder can ship an MVP in days, not months. A non-technical founder can ship in weeks. The cost of testing an idea dropped from $500K (team + time) to approximately $5K (AI tools + time). More experiments → more power-law draws → more potential outliers in the ecosystem.

2. The pre-seed stage is compressing. If you can build a working product before raising, the traditional pre-seed pitch deck becomes a demo. VCs who still evaluate slide decks as proxies for execution ability are screening a shrinking pool. The founders who can't yet show a demo are a smaller and smaller share of the fundable universe.

3. AI tools are themselves the biggest power-law winners of this era. OpenAI ($25B ARR), Anthropic ($19B ARR), Cursor ($2B ARR) are growing faster than any venture-backed companies in recorded history. ChatGPT went from $0 to $25B ARR in approximately 3 years. This raises a genuine question for fund managers: should your thesis include direct exposure to AI infrastructure, or only to AI-enabled applications?

4. Fund operations compress too. AI screening (NUVC), AI due diligence, AI memo generation, AI portfolio monitoring. The $120K analyst is being replaced by $199/month software at the task level. This changes fund economics for emerging managers — particularly sub-$50M funds where operating costs were previously prohibitive.

5. The OpenClaw signal. When an open-source AI agent framework accumulates 250K GitHub stars in 4 months (more than React accumulated in 11 years), the developer ecosystem is signalling that agentic AI is the next platform layer. VCs who understand agent-native architecture will back the next cycle of power-law companies. VCs who don't will be screening the same deal flow as everyone else.

A Framework for Evaluating AI-Era Companies (Without Needing NUVC)

Four questions to apply to any AI-adjacent investment thesis:

1. Is the moat capability or network? Network-effect moats are established (social graphs, marketplaces). Capability-effect moats are still being competed for. A company with a proprietary dataset that improves its model is building a capability moat. A company that is only using OpenAI's API is not.
2. Is the adoption curve being driven by individual utility or organisational deployment? Individual adoption is faster (Claude Code hit $1B ARR on individual developer subscriptions). Organisational deployment is stickier (GitHub Copilot's 90% Fortune 100 penetration). Best-in-class companies do both.
3. What is the marginal cost of serving the next user? AI tools approach zero marginal cost at scale. If your company's cost structure does not benefit from scale (i.e., each new customer requires proportional human service), the adoption curve compression works against you, not for you.
4. Is the team using AI internally to build, or are they building AI products? NUVC is an example: two non-technical founders built a 340,000-line platform with AI agents — no CTO, no engineering team. See Two Founders, 19 AI Agents, Zero Employees. Teams that use AI for their own operations compound the productivity advantage of the tools they build.

See also: Why VCs Are Using AI Wrong and The One-Person VC Fund.

What Does This Mean for Founders Raising in 2026?

Three things have changed permanently:

• Your competitive benchmark is global and faster. The 2-month adoption curve of ChatGPT means VCs have seen what extreme product-market fit looks like. A strong Australian seed company growing 15% month-on-month is competing for attention with global comparables growing 30%+ in AI categories.
• Demo beats deck. If you have built with AI tools, show the product. The adoption speed of AI developer tools means any competent team can ship a prototype now. If you haven't, that absence is itself a signal.
• Your operating cost advantage is a feature. A founding team using AI tools to replace $300K of headcount is running at a different margin profile than a comparable team from 2021. Model this explicitly in your pitch — it's a real structural advantage.

If you are a founder, get your NuScore to see how your pitch deck is read by AI-native investors. If you are a fund manager building an AI-native process, NUVC's investor platform is the infrastructure layer.

Frequently Asked Questions

How fast did ChatGPT reach 100 million users?

ChatGPT reached 100 million users in approximately 2 months after its launch in November 2022, making it the fastest consumer application to reach that milestone at the time. As of March 2026, ChatGPT has 800 million weekly users and $25 billion ARR (OpenAI). For context, Instagram took 2.5 years to reach 100 million users, TikTok took 9 months, and Facebook took 4.5 years. The acceleration from TikTok (9 months) to ChatGPT (2 months) represents a 4.5x compression in adoption speed.

How does AI adoption compare to social media adoption?

Social media adoption is driven by network effects — each new user adds value for existing users by expanding the social graph. AI adoption is driven by capability effects — model improvements benefit all users simultaneously regardless of user count. This means AI tools can achieve immediate utility for isolated users without waiting for a network to build. The result is faster initial adoption (ChatGPT, 2 months; OpenClaw, 48 hours to 100K GitHub stars) but different retention dynamics — retention depends on model improvement speed rather than social graph lock-in.

What is a capability effect versus a network effect?

A network effect means each additional user increases the product's value for existing users (classic examples: telephone, Facebook, marketplaces). A capability effect means improvements to the underlying technology increase value for all existing users simultaneously, without requiring more users. AI platforms like ChatGPT, Claude, and DeepSeek grow through capability effects — when the model improves, every user benefits instantly. This creates a fundamentally different competitive moat: the company with the best model or the best proprietary training data has a structural advantage that compounds with investment, not with user acquisition spend.

How does faster technology adoption change venture capital?

Faster adoption curves change venture capital in three ways: (1) the time from investment to proof-of-product-market fit compresses, meaning early-stage conviction is more valuable relative to growth-stage follow-on; (2) build costs have dropped 10x with AI developer tools, meaning capital efficiency metrics look different — a $5M seed round that would have funded 12 months of an engineering team in 2021 can now fund 3-4 years of AI-assisted development; (3) the categories generating the largest returns are shifting — the fastest-growing companies of 2023-2026 (OpenAI, Anthropic, Cursor) are AI infrastructure companies, not AI-enabled application companies, which has implications for where fund managers focus their theses.