OpenAI is booming – and under siege

OpenAI has become the shorthand for the AI boom: headline valuations, explosive user growth, and products that feel like science fiction. Business Insider recently framed the moment starkly: a company reportedly worth around $300 billion with hundreds of millions of users – and sharks circling from every direction. That framing captures the tension of 2024 and beyond: scale and momentum on one side, legal fights, regulatory probes, and fierce competition on the other. (Business Insider)

This article unpacks the two big numbers you keep seeing – valuation and users – and adds the context that matters: revenue realities, regulatory heat, lawsuits, partner dependencies, and the moves OpenAI is making to keep its lead.

What does a $300 billion valuation really mean?

Private company valuations are often set by tender offers and secondary sales, not public markets. In early 2024, OpenAI was reported to be exploring a share sale that would value the company at roughly $80-90 billion. Later reports and investor marks suggested the number could be far higher, with some speculation creeping toward the $150-200 billion range, and more recent chatter pushing as high as $300 billion. Treat those top-end figures as directional, not definitive. (WSJ) (Bloomberg)

Behind the valuation is real revenue growth. Reporting in mid-to-late 2024 indicated OpenAI had reached an annualized revenue run rate of roughly $3.4 billion, up dramatically from the prior year, driven by API usage, ChatGPT Enterprise, and licensing deals. (The Information) (Reuters)

Costs are the counterweight. Training and serving frontier models are capital intensive, with compute, data center, and energy costs rising. The industry is still supply-constrained on top-tier AI chips, even as availability improves. (Financial Times)

Sam Altman has publicly explored radical ideas to secure long-term compute supply, including conversations about multi-trillion-dollar investments to scale semiconductor manufacturing capacity. Ambitious as that sounds, it underscores a hard reality: competitive advantage in AI is increasingly a function of access to compute and energy. (WSJ)

Do 500 million people really use ChatGPT?

The precise number depends on how you count. OpenAI itself disclosed 100 million weekly active users in November 2023, a rare hard metric. (OpenAI DevDay) Independent traffic firms have since estimated billions of monthly visits to chatgpt.com and the mobile apps, implying a very large base of recurring users – potentially in the hundreds of millions – though exact monthly active user figures are not publicly confirmed. (Similarweb)

Enterprise adoption is a major tailwind. OpenAI’s products are embedded in Microsoft Copilot and Azure OpenAI Service, used by thousands of enterprises for coding assistance, knowledge retrieval, and customer support. This distribution matters as much as raw consumer traffic. (Microsoft)

Bottom line: the user base is enormous and still growing, but be skeptical of precise MAU claims unless they come directly from OpenAI.

Why the sharks are circling

1) Big Tech competition

• Google is pushing Gemini and long-context models into Workspace and Android, competing head-to-head in consumer and developer markets. (Google)
• Meta is seeding open models like Llama 3, lowering the cost of building capable AI into products and potentially eroding proprietary model premiums. (Meta)
• Anthropic’s Claude 3 family has earned strong reviews for reasoning and safety, appealing to enterprises with a reliability-first pitch. (Anthropic)
• Apple announced Apple Intelligence, with on-device models and a selective integration with ChatGPT for certain complex queries. Great distribution – but Apple will keep many tasks in its own ecosystem. (Apple)

2) Legal pressure from rights holders

• The New York Times sued OpenAI and Microsoft in December 2023, alleging copyright infringement from training and outputs. The case could set precedents for AI training on copyrighted text. (NYT)
• In parallel, OpenAI has pivoted to licensing content at scale, striking deals with major publishers and platforms, including News Corp, Reddit, and Stack Overflow. These deals legitimize training data access and reduce legal risk. (News Corp) (Reddit) (Stack Overflow)

3) Regulators are awake

• U.S. regulators have divided oversight: the FTC is probing Microsoft’s relationship with OpenAI while the DOJ looks at Nvidia, reflecting antitrust concerns across the AI stack. (NYT)
• The FTC separately opened an investigation into OpenAI’s data practices in 2023, centering on privacy and security. Expect continued scrutiny. (Washington Post)
• In Europe, the EU AI Act sets a risk-based regulatory regime. Frontier model providers will face transparency, safety, and governance obligations, with fines for violations. (European Parliament)

4) Safety, governance, and talent

• High-profile departures in 2024 – including Jan Leike and Ilya Sutskever – sparked debate about how OpenAI balances rapid productization with long-term safety. (The Verge)
• OpenAI responded by forming a Safety and Security Committee to review and strengthen internal processes and risk management. (OpenAI)
• Controversy over ChatGPT’s voice assistant – particularly the resemblance of a voice named Sky to actress Scarlett Johansson – highlighted reputational risks in product design. OpenAI paused that voice after concerns. (BBC)

The Microsoft factor: superpower and single point of failure

Microsoft is OpenAI’s most important partner and investor, providing capital, distribution, and – critically – cloud compute via Azure. The partnership has enabled OpenAI to scale quickly and has put OpenAI models in front of hundreds of millions of Microsoft users through Copilot. (Microsoft)

But dependence has risks. Antitrust scrutiny now explicitly examines how Big Tech partnerships may entrench power in AI. And strategically, OpenAI must keep innovating to remain Microsoft’s best-in-class supplier while also maintaining product independence such as ChatGPT and the API business. (NYT)

Strategy in motion: how OpenAI is defending the lead

Product velocity

OpenAI’s cadence remains fast. Releases like GPT-4o pushed toward multimodal, low-latency experiences, and the company continues to improve reasoning, tools, and enterprise controls. Model quality and cost curves – not flashy demos alone – will decide whether OpenAI stays ahead as rivals close the gap.

Data legitimacy through licensing

After a year of lawsuits and public debate over training data, OpenAI has moved decisively toward content licensing. Deals with major publishers and platforms reduce legal uncertainty and improve freshness and reliability of model answers. Expect more agreements as publishers seek revenue and attribution. (News Corp) (Reddit) (Stack Overflow)

Distribution everywhere

From ChatGPT Enterprise to integrations in Microsoft 365 and selective use inside Apple’s ecosystem, OpenAI’s strategy is to be wherever people work and create. Enterprise contracts, developer APIs, and platform deals provide multiple paths to monetize the same core models. (Apple) (Microsoft)

Compute and energy pragmatism

Securing affordable, reliable compute is existential. Expect OpenAI to keep deepening ties with Azure, explore custom hardware partnerships, and support energy innovations that shrink the cost to serve. This is the quiet backbone of the AI race.

What to watch next

• Legal outcomes that shape training data norms – especially the NYT case and other copyright suits.
• Antitrust actions testing the Microsoft-OpenAI relationship and the AI stack more broadly.
• Model breakthroughs in reasoning and efficiency, including long-context and agentic capabilities.
• Hardware and data center announcements that signal how compute bottlenecks will be relieved.
• More publisher and platform licensing deals that set pricing baselines for data.

Bottom line

OpenAI sits at the center of AI’s biggest contradictions: breathtaking adoption and escalating costs, market power and regulatory pressure, open research roots and a hard pivot to enterprise-grade products. Whether the company is worth $90 billion or $300 billion, the strategic imperatives are the same: keep the models best-in-class, secure compute and energy at scale, de-risk data, and earn trust with users, partners, and regulators. The sharks are real – but so is the lead.

FAQs

Is OpenAI really worth $300 billion?

Reports vary widely. Private tender offers in 2024 pointed to $80-90 billion, with some later investor marks speculating far higher. Treat any $300 billion figure as unconfirmed. (WSJ) (Bloomberg)

How many people use ChatGPT?

OpenAI disclosed 100 million weekly active users in November 2023. Third-party traffic data suggests hundreds of millions of users may interact with ChatGPT monthly, but exact MAUs are not public. (OpenAI) (Similarweb)

What are OpenAI’s biggest legal risks?

Copyright suits led by The New York Times challenge how AI models can be trained on copyrighted text and how outputs are handled. Outcomes could reshape licensing norms and model design. (NYT)

How dependent is OpenAI on Microsoft?

Very. Azure provides core compute, and Microsoft integrates OpenAI models in Copilot and across its products. The relationship accelerates OpenAI’s reach but draws antitrust oversight. (Microsoft) (NYT)

What does the EU AI Act mean for OpenAI?

Expect transparency, safety, and governance requirements for frontier models, plus potential audits and fines for violations. Compliance will add cost but could also raise the bar for the entire market. (European Parliament)

Sources

1. Business Insider – coverage of OpenAI valuation and user growth
2. OpenAI – DevDay 2023: 100 million weekly active users
3. Microsoft – multiyear, multibillion-dollar OpenAI partnership
4. The New York Times – lawsuit against OpenAI and Microsoft
5. The New York Times – FTC and DOJ antitrust inquiries into AI leaders
6. European Parliament – EU AI Act overview
7. News Corp – OpenAI content licensing partnership
8. Reddit – OpenAI partnership announcement
9. Stack Overflow – OpenAI partnership announcement
10. The Verge – OpenAI safety leadership departures
11. OpenAI – establishing a Safety and Security Committee
12. Apple – Apple Intelligence and ChatGPT integration
13. BBC – OpenAI pauses Sky voice after Scarlett Johansson concerns
14. Financial Times – AI chip supply constraints context
15. Wall Street Journal – Sam Altman explores trillions for AI chips