The business in AI: trivial to transformationalNew 

The next five years will be decisive in shaping the future of a business in AI, with one caveat: the government as a customer. The recent stand-off between Anthropic (essentially focused on enterprise customers) and the Pentagon (US Federal Government institution) and its possible impact on individual customers is a potential factor. This became clear when OpenAI received a lot of flak for stepping into the vacancy created by the exit of Anthropic.

A different kind of platform

Gemini has an advantage over OpenAI as both pursue individuals as customers, because Google’s ‘platform’ is dramatically different from Apple’s walled system: According to a LinkedIn post, Android reaches 3 billion devices, Google Search processes 8.5 billion daily searches, and Google Workspace serves hundreds of millions of business users, as the LinkedIn post notes. Its 8 million enterprise users are spread over 2,800 enterprises, with API (application programming interface) calls more than doubling to 85 billion!.

Google Gemini’s sponsorship’s deal with BCCI for Tata IPL 2026 is for Rs 260 crore is small change for Google not just in terms of absolute spend but more important in terms of what it will fetch Gemini. The advertisements, during every IPL match, featuring the different purposes for which Gemini can be used, clearly demonstrates that it is an entire world of trivial uses, which would normally have been satisfied by resorting to a search engine such as Google itself. According to FICCI-EY report, IPL 2025 reached 1.18 billion viewers across television and digital platforms, with an increase this year. Even if 10% of viewers become Gemini customers, it translates into hundreds of millions of dollars.

Google Gemini earned $1.2 billion in 2025 from subscriptions; it earned $300 million 2024. It had 90 million users in 2024 and now boasts of 750 million users as of December 2025.  

The standalone fight: OpenAI

Reuters (January 19, 2026) quoted OpenAI CFO Sarah Friar from her blogpost saying that OpenAI’s annualised revenue crossed $20 billion (which means that it did not actually earn $20 billion but calculating annual revenue based on actual revenue earned during a part of the year).  It also quoted their Chief of Policy, Chris Lehane, who said that OpenAI will release its device during the second half of 2026. On OpenAI’s website, Friar writes that “Compute grew 3X year over year or 9.5X from 2023 to 2025: 0.2 GW in 2023, 0.6 GW in 2024, and ~1.9 GW in 2025. While revenue followed the same curve growing 3X year over year, or 10X from 2023 to 2025: $2B ARR in 2023, $6B in 2024, and $20B+ in 2025”. Compute is a product of OpenAI.

The future – enterprise and individuals

According to another post on the OpenAI website by Friar, OpenAI raises $122 billion to accelerate the next phase of AI, claiming that “OpenAI is becoming the core infrastructure for AI, making it possible for people around the world and businesses, big and small, to just build things”. In February, 2026, Reuters reported that OpenAI sees a compute spend $600 billion by 2030.  

According to Enlyft, “We have data on 196,248 companies that use OpenAI. The companies using OpenAI are most often found in United States and in the Information Technology and Services industry. OpenAI is most often used by companies with 1-10 employees and 1M-10M dollars in revenue. Our data for OpenAI usage goes back as far as 3 years and 2 months”. According to them, Information Technology and Services (15%), Computer Software (11%), Marketing and Advertising (8%) and Internet (7%) are the largest segments. Its market share is 69%.

Google Gemini – 35,463 companies, most often found in United States and in the Information Technology and Services industry, and used by companies with 1-10 employees and 1M-10M dollars in revenue. Enlyft’s data for Google Gemini usage goes back as far as 2 years and 3 months. It has  a 11% market share.

The nearest competitor of OpenAI is Anthropic, which was in the news for its confrontation with the Pentagon, with Claude as its leading product. The reason why Anthropic is not a household name is that it is a business-focused company; their main product ‘Claude’ is a ‘work tool’ for business users. As anyone even reasonably familiar with the business of IT would know, enterprise is a vastly different business compared to the universe of individuals.

API, API integration and AI

One of the major parts of the enterprise AI business is API integration. Every software system will have a set of APIs – application programming interface, indicating that the many applications that a system has, need to communicate with one another to exchange data, features and functionality; the system will not work otherwise. Since every system has many applications (not necessarily from the same company), interface management is a task by itself. Thus not only do we need API, we need API integration, just as we needed system integration. Now AI is capturing this market. This is a logical corollary of enterprise users taking to AI.

The global AI API market is projected to register a CAGR of 32.2% during the forecast period, reaching $179.14 billion by 2030 from an estimated $ 44.41 billion in 2025. Some leading players in the AI API market include Google, Microsoft, IBM, OpenAI, AssemblyAI, Hugging Face, DeepSeek (China), etc.

A boutique or the specialist?

Clearly, there is a huge and growing distinct businesses.

Will the enterprise market be dominated by enterprise-specific companies? Will companies who target both enterprises and individuals sustain their initial victories in the enterprise segment? What will sway investors? This is an important question because minimum investments are in the $10 billion plus category. We will just have to wait and watch.

There is just one guarantee: AI will not ‘take over’ mankind! There is just a lot of money to be made!