The $115 Billion Question: What Two AI Giants' Combined ARR Really Tells Us About Centralization

There is a number floating through the crypto and tech press this week that should stop you cold: $115 billion. That is the combined Annual Recurring Revenue reportedly achieved by OpenAI and Anthropic in 2026. Not projected. Not targeted. Achieved. Before you let that figure wash over you as another bullish headline in a market hungry for good news, I want you to sit with what it actually means. Because tracing the code back to the conscience behind it, this isn't just a story about two companies getting rich. It's a story about the concentration of power in the AI layer, and what that means for the rest of us building in the decentralized stack. The report originated from Crypto Briefing, a publication more accustomed to covering token prices than enterprise software metrics. That alone should raise a flag. But let's assume the number is directionally accurate. If OpenAI holds roughly $800 billion of that ARR and Anthropic the remaining $350 billion, we are looking at two entities that have each individually surpassed the annual revenue of Salesforce, Oracle, and SAP. These are no longer startups. They are infrastructure. And the speed at which they got here—the article explicitly uses the word 'accelerates'—suggests we have crossed a threshold that the broader tech world has not yet fully processed. I spent four months in 2017 auditing ERC-20 standards for three ICO projects in Cape Town. I found critical reentrancy vulnerabilities in two of them that later collapsed, saving a small group of investors roughly $45,000. That experience taught me that technical precision is a form of social protection. And when I look at this $115 billion figure through that same lens, I see something uncomfortable: the market is rewarding speed of deployment over robustness of architecture, and the entire industry is following suit. Let's break down what this ARR actually implies. At a 10-20x SaaS multiple, we are talking about a combined valuation between $1.15 trillion and $2.3 trillion. That puts OpenAI and Anthropic in the same weight class as Apple and Microsoft. The market has moved from narrative-driven speculation to revenue-driven valuation. That is real progress. But it also means the stakes for failure have multiplied exponentially. Every line of code is a hand extended in trust, and when you are processing trillions of tokens daily, a single alignment failure or security breach doesn't just hurt a company—it erodes trust in the entire AI ecosystem. Here is where my contrarian instinct kicks in. The conventional reading of this news is that AI is winning, and by extension, the technology sector is healthy. I see something different. I see a duopoly forming that will make the centralized web look positively diverse. The report notes that combined, these two companies may control 40-55% of the global AI software market. That is not a market. That is a chokepoint. And in my 2021 work with indigenous South African digital artists on NFT royalty enforcement, I learned what happens when platforms control the pipes: creators get squeezed. Artists own their pixels; we just hold the keys. But when a handful of companies control the inference engines that generate those pixels, the keys might as well be made of wax. The deeper problem is what this ARR does to the incentive structure for open-source development. I have been an open source evangelist for over a decade, and I believe with my whole chest that open source is not a license; it is a promise. A promise that the community can audit, fork, and improve the tools we all depend on. But when $115 billion in annual revenue is on the line, the pressure to keep models behind APIs, to restrict weights, and to gate access behind enterprise contracts becomes almost irresistible. The tragedy is that this is happening at the exact moment we need transparency most. The EU AI Act and similar regulations are struggling to keep pace, and the compliance costs they impose will disproportionately hurt the small projects that cannot afford armies of lawyers. Let me be clear about what I am not saying. I am not arguing that OpenAI and Anthropic are malicious actors. They have both demonstrated genuine commitment to safety research, and Anthropic's focus on interpretability is genuinely commendable. My concern is structural, not personal. When revenue reaches this scale, the organization's primary obligation shifts to shareholders and growth targets. The 2022 bear market taught me that resilience is built in advance, not during the crash. And the same logic applies here: we need to build the decentralized alternatives now, while these companies are still relatively benevolent, because the window for meaningful competition will close faster than most people expect. The infrastructure implications alone should terrify anyone who cares about decentralization. At 20-35% of revenue going to compute, we are looking at annual spending of $230-345 billion on GPUs and cloud services. That level of demand gives these companies enormous leverage over chip suppliers and cloud providers. It also means that a single supply chain disruption could ripple through the entire AI economy. I have been tracking the convergence of AI and decentralized identity protocols since 2025, and I can tell you that the solution to this concentration problem is not technical—it is philosophical. We need to build systems that assume centralization will happen, and design for graceful degradation and community sovereignty from day one. Education is the only true decentralized currency, and that has never been more relevant than it is today. The knowledge gap between those who understand these systems and those who simply consume them is widening. When I ran my 'DeFi for Everyone' workshops in Cape Town during the 2020 summer, I saw firsthand how education empowered people to make better decisions with their capital. The same principle applies to AI literacy. The more people understand how these models are trained, deployed, and monetized, the more pressure they will exert for accountability and transparency. So what is the takeaway from this $115 billion milestone? It is not that AI is unstoppable, nor that we should all abandon hope. It is that we are entering a new phase where the battle is no longer about capability—it is about custody. Who holds the keys to the most powerful technology ever created by humanity? Who decides what questions it will answer and what tasks it will perform? These are not technical questions. They are moral ones. We build bridges, not just blocks, between people, and that means we must start constructing the governance frameworks and decentralized alternatives today, while we still have the luxury of choice. The code is writing itself. The question is whether we will have the conscience to direct it.