Sam Altman is panicking...

Kind: captions Language: en Right now, 92% of the Fortune 500 runs on OpenAI software. Last year, OpenAI lost more than $5 billion, and the losses are still mounting. [music] Yet, corporate America is becoming too tied to OpenAI to simply walk away. It's a 2008 style financial hostage situation. And if Sam Alman gets his way, you're the one who will have to pay the ransom. [music] The world's most successful firms could have waited to see how the AI industry would evolve. They could have shown patients, monitored the market, and watched and waited to see which AI models would prove the smartest and most efficient. Instead, almost all of them jumped on board the OpenAI bandwagon. Some were caught up in the hype. Others were fearful of being left behind. The Fortune 500 world moves fast, and when they see what they think is a golden opportunity, they grasp it with both hands. That's what happened in August 2023 when OpenAI effectively introduced AI to the working world with the release of Chat GPT Enterprise. Within a matter of months, over 80% of Fortune 500 firms had embraced the company software. By mid 2025, that had increased to 92%. OpenAI had more than 600,000 business clients across major industries. Soon, the companies were hooked. Many started off using it on an experimental basis for small tasks and projects, and it quickly built up toward larger and more serious applications and companywide deployments. Today, large parts of those firms operations aren't just AI powered, they're AI dependent or open AI dependent. Sam Alman's company went from a relative unknown to a household name in just a couple years. It infiltrated the world's most important industries and found favor among the globe's biggest brands. But there is another side to the story, one that paints it in a far more concerning [music] light. Because as well as being an unprecedented market victory for OpenAI, the 92% dependency is also a serious vulnerability. The entire American economy has in effect placed its fate in the hands of just one provider. A single point of failure, one that is bleeding out at breakneck [music] pace. In 2024, OpenAI generated $3.7 billion in revenue. It also spent [music] 12.4 billion for a net loss of 5.09 billion. In 2025, it got significantly worse. The company [music] made 13 billion in revenue, but spent $34 billion. It also suffered $20 billion in operational losses, leading to a net loss of more than 38 billion. That's like losing the market cap of a mid S&P 500 firm like PayPal or United Airlines in the span [music] of just 12 months just to keep the lights on. Open AAI is massively in the red. It is a failing business being asked [music] to prop up the world's biggest economy and that is a recipe for disaster. But how did it reach this point? On the surface, it seemed as though OpenAI was making all the right moves to achieve total market dominance. Its flagship product, Chat GPT, quickly became more popular than any alternative on the market. It was the go-to choice for hundreds of millions of users across the globe and the fastest growing web platform of all time. Slowly but surely, competitors began to emerge. Gemini, Claude, Perplexity, and Deepseek. They proved that the AI industry wasn't going to be dominated by a single company. And in their own ways, some of those rivals began to outperform OpenAI's products as well. They began to erode Chad GPT's monopoly. OpenAI's market share also fell from a dominant 80% to as low as 46% by the middle of 2026. Political backlash also hurt OpenAI's popularity with uninstalls spiking after its agreement with the US Department of Defense. With big money flowing into other AI tools, the OpenAI monopoly was effectively over as quickly as it began. But the single biggest issue facing OpenAI today, the one that threatens the very future of the company as we know it, is the rise of open- source alternatives to its premium AI products. 2026 in particular, has seen a tidal wave of companies transitioning from renting AI solutions to [music] building their own AI capabilities thanks to open-source solutions. It's not hard to understand this shift. The benefits of opting for open-source are obvious. Open source users have more flexibility. They're able to pick and choose between a wide range of models. Companies are free to experiment and flex their creative muscles to gain a competitive edge in their respective markets. But [music] the biggest and most important advantages are the cost benefits. Renting AI solutions can be expensive. Particularly for larger businesses that scale their operations over time. Companies can easily spend six even figure sums on OpenAI products every single year. That is just for the rudimentary tasks like AI chats and content summarization workflows. Open- source models, however, don't come with those high fees. That doesn't mean they're completely free either. There are costs involved with self-hosting AI models, including the infrastructure required to run them. But these fees are significantly less than what companies would spend on OpenAI for credits and subscriptions. Models like Meta's own Llama 3.145B can now be self-hosted for a fraction of the cost of OpenAI's API fees. So why would anyone continue paying OpenAI's prices? They wouldn't, [music] and they're not. Increasingly, companies are shifting away from OpenAI and other premium AI models and into the open-source market. Around half of all Fortune 500 companies now utilize open-source AI. Clint Dong, CEO of HuggingFace, an open- source AI platform, believes this could change the AI landscape in the future. He predicts the big brand frontier models will be used for experimenting and some really high-v valueue tasks while most of the production workloads will actually be powered either by private models within companies or by open-source models. The long argues that companies increasingly want to feel in control of the AI. Microsoft CEO Satya Nadella agrees and has warned companies not to get locked in with a single provider, urging them instead to control their own learning loop. These words might not mean much to OpenAI and Sam Alman if they were coming from the head of a random company, but from the head of Microsoft itself, they mean everything. Microsoft has been OpenAI's biggest backer since 2019. That's when a deal was struck that involved the Windows creator providing a billiondoll injection to help OpenAI speed toward artificial general intelligence or AGI. In return, Microsoft enjoyed exclusive status as OpenAI's cloud provider as well as exclusive rights to license and monetize the company's pre-AGI technologies. [music] That agreement has since been drastically restructured, but the tech giant has poured well over $100 billion into the AI company. It remains OpenAI's largest shareholder with a 27% stake as of June 2026. While their relationship has changed over the years, the two firms have long been seen as solid partners. In many ways, OpenAI owns a large part of its popularity and existence to Microsoft. It might have other shareholders and investors, but the Chat GPT creator still relies heavily on Microsoft's infrastructure and enterprise deployment to remain operational. That's about to change because just like hundreds of other Fortune 500 firms, Microsoft is beginning to pivot away from its longtime partner. According to Bloomberg, the company has already started the process of replacing OpenAI models in Microsoft 365 applications with its own proprietary MAI solutions. Tens of thousands of weekly prompts that were once handled by OpenAI are now being handled by Microsoft's in-house models. It's still a small portion of the company's overall AI usage, but it is just the start of something that is likely to grow in scale and scope in the years to come. Microsoft isn't keeping that strategy under wraps. It is openly owning its intentions to move away from frontier models and focus more heavily on proprietary alternatives. At the annual Microsoft Build developer conference in June 2026, the company's AI CEO Mustafa Sullyman unveiled seven new MAI models. These aren't just entry-level tools. They are highly intelligent, highly capable AI models. MAI Code 1 Flash can reportedly match Anthropic's Opus 4.6 model for performance while offering a lower operating cost. And even to one of the wealthiest companies in the world, costs matter. For Microsoft, every interaction with Copilot, which is powered by OpenAI, costs tokens and consumes GPU capacity, memory, and storage. Companies need to look for ways to keep their cost per request as low as they can. For Microsoft, that means moving away from OpenAI and other AI providers. Microsoft is building its own life raft to escape Sam Alman's sinking ship. Once the biggest backer leaves OpenAI's financial future will go from bad to worse, far worse. But the powers that be don't want that to happen. They don't like the thought of a future in which open- source AI models replace premium providers. It's good for the public, sure, but not for investors. They need big brands like OpenAI and Anthropic to continue to dominate the AI landscape in order to make their millions. And that's why there's been strong lobbying movements in the halls of Washington to support legislation that could criminalize the competition. Much of this legislation is cleverly written to appear like a net positive, altruistic even. But many bills and proposals conceal secret loopholes and hidden agendas that could make life much easier for the likes of OpenAI. At the same time, it becomes harder for smaller providers and open- source communities that challenge them. California's Safe and Secure Innovation for Frontier Artificial Intelligence Models Act, or SB1047, attempted to bring new rules to the AI industry. The 2024 bill would have required major developers to test models for catastrophic risks before launch, sparking a major debate over AI safety and regulation. It stated that if passed, it would apply to the largest models, those costing more than $100 million to train. It mandated that developers build technical mechanisms or kill switches to completely shut down their AI systems in emergency situations. It would also provide legal protections for whistleblowers who report safety violations or reckless behaviors. It all sounds fair. A clear attempt to hold the biggest AI brands accountable for their actions and their products. But there's another side to it. What initially looked like an altruistic quest for AI safety is just another desperate and aggressive corporate strategy to outlaw free competition. In the words of the Reason Foundation think tank, it would criminalize the development and use of open-source AI models. Reading between the lines, experts realized that the strict regulations involved in SB 1047 could stifle innovation and consolidate market power amongst big tech. Companies like OpenAI would be in stronger positions than ever before. Why? Critics argued that only the biggest AI companies could afford the legal teams, compliance systems, and audits required by SP 1047. Smaller developers in open source projects, they warned, could be locked out of the market entirely. Much of the open source ecosystem involves small teams and independent researchers building and iterating on top of existing models. SP 1047 would have made it so that the original developers of those models like OpenAI would be held liable for any downstream modifications made by third parties. As a result, they would have been forced to stop releasing open- source models to protect themselves against costly, damaging lawsuits. And while there were positives associated with SB 1047, the bill could have slammed the gates on the open source community and handed the keys to Sam Alman. Even though Alman opposed the bill in public, he wouldn't have been upset if it had passed. But that didn't happen. So the OpenAI boss began to implement a bigger, more complicated strategy, one that could impact American taxpayers for generations to come. In order for this plan to work, Altman needed to convince the world that OpenAI was indispensable. He had to prove not just to partners and shareholders, but to the United States government that his company was too big and too important to ever be allowed to fail. The first step of this master plan was OpenAI's late 2025 restructure into a public benefit corporation or PBC that effectively unlocked massive amounts of capital to fund the company's extensive AI research and infrastructure projects. It also removed pre-existing profit caps in order to provide greater incentive for investors and shareholders to back the brand. The restructure paved the way for the largest private funding round in recorded history. In March 2026, tech giants and investors poured a combined 122 billion into OpenAI. Since then, despite its miserable net losses, OpenAI has soared to an utterly absurd valuation of around $852 billion, around 35 times its current annualized revenue. It is an insane amount of money. But Altman had much more in mind. Since 2024, he's been trying to raise $7 trillion to reshape the future of AI infrastructure. This ambitious plan would supercharge chip manufacturing, strengthen global supply chains, and build the data centers needed to power advanced AI models. $7 trillion is around 6% of the entire global GDP. It is more than the entire economic output of entire continents like South America and Africa. It is a staggering amount of capital, one that does not exist in the private market. And that's why Alman has spent years searching for backers. Yet, no government, sovereign wealth fund, or major consortium has committed to funding the vision. It might look like the OpenAI CEO is failing. He's not. He knew all along that he was never going to get anyone to agree to provide the investment. Whether the money materialized or not didn't really matter. What mattered were the numbers. An $852 billion valuation, a $7 trillion investment plan. A company that's associated with that kind of cash sounds like it must be one of the biggest and best in the world. Nobody would think a CEO was going around asking for trillions of dollars from the United Arab Emirates. If his company was making nearly a $40 billion loss, it wouldn't make sense. And that's the trick. The numbers are an illusion, one that's been carefully crafted by Altman to make OpenAI seem so much more valuable and influential than it actually is. So even as the company's losses mount, the market share collapses and partners back away from it, it still seems like the number one name in the industry. And that leads us to the culmination of Altman's grand plan. On July 2nd, 2026, numerous news agencies reported that the OpenAI CEO had closed door meetings with the United States government. He supposedly offered the administration a 5% stake in the company. Altman reportedly argues that by giving the government and by extension the American public a financial stake in the company, it's the perfect way to spread the wealth of AI. It would also be the bailout his company desperately needs to survive. It was easy enough for OpenAI to coast along when it was the only name in town, but with rivals gaining ground and enterprise clients actively investigating open- source alternatives, Altman is panicking. He needs to grab any lifeline he can find. A government deal would be the ideal lifeline and the ultimate Trojan horse. But if and when the US Treasury takes part ownership of OpenAI, the government is no longer just a regulator, but a shareholder. And from that point on, any government decision affecting AI, from regulation to infrastructure spending, could impact the value of a government-owned asset. Even if officials claimed to remain impartial, critics would question whether they could really make unbiased decisions while holding a significant financial stake in the industry. They'd have a greater incentive than ever before to take actions that benefit OpenAI. They could implement controls to limit the influence of foreign competitors or draft federal procurement plans to incorporate OpenAI's products into the federal digital ecosystem. If the situation continued to deteriorate for OpenAI, Washington could even step in with emergency funding to keep the company afloat. Taxpayers would be taken along for the ride, whether they want it or not. Their money would become tied to OpenAI's financial future. The government responsible for protecting public interests could face pressure to make decisions that benefit the company instead. And equity doesn't expire. This wouldn't be a temporary loan or a contract. It would be a long-term stake that could carry on from administration to administration. Over time, that could push OpenAI closer to becoming too strategic to fail. Not because it's the best company in the market, but because its success would be connected to public assets and national infrastructure. Sam Alman promised to change AI. Now the world has to pay the price. People are turning their backs on Open AI, but one breakup hurts more. Watch massive Microsoft divorce that will bankrupt Open AI and ChatGpt forever to see where this leaves the AI giant. Or click this video instead.