openai

OpenAI employees who sold their shares

by

In October of last year, OpenAI closed a secondary share sale which raised its valuation to 500,000 million dollars (Today it is already worth 852,000 million). This allowed employees to sell their shares, becoming multimillionaires even before the IPO. 6.6 billion. It is the total amount of the operation in which more than 600 employees, both current and former employees, benefited. In previous similar operations, OpenAI limited the maximum per person to 10 million, but in this case, due to high demand from investors, they decided to triple it to 30 million. Of all of them, 75 employees reached the maximum number, becoming multimillionaires in one fell swoop. The AI ​​winners. Uncertainty about the future profitability of AI continues to loom large, but that is not affecting workers in the most important AI laboratories. The case of the secondary sale of OpenAI is just one example of how AI engineers have become the biggest winners of this boom. Last summer, Meta offered up to $100 million to competing engineers and NVIDIA paid 900 million by an employee. Tender offers. The usual thing when you started to work in a startup is that you received a low salary and a lot of shares, but you had to wait for the IPO to be able to make cash. This made many employees rich only on paper, but without real liquidity. A tender offer allows employees to get paid much sooner, allowing them to sell stakes to private investors. They count in the Wall Street Journal That this mechanism, which was previously a one-time thing, has become a central piece in Silicon Valley achieves a double effect: in addition to turning employees into millionaires in advance and thus retaining them to stay in the company, it helps to consolidate stratospheric valuations, causing each new operation to set a higher reference price for OpenAI shares. The local impact. The rain of millions had an almost immediate consequence on the real estate market in San Franciscowhich is seeing prices rise even more. In February of this year the rents had increased by 14% compared to the same period in 2025 and the purchase prices of apartments and single-family homes rose by 12 and 23% respectively. At the same time, the sale of homes valued above $5 million has increased by 220%. The AI ​​gap. In the end, the AI ​​boom is not only redefining which companies rule Silicon Valley (and the world), but also who can afford to live there. The combination of exorbitant valuations, tender offers billionaires and a stressed real estate market is turning AI engineers into a new urban aristocracy that, in practice, is redefining what it means to have a “good salary.” The income that a few years ago guaranteed access to the best neighborhoods is no longer enough today. Image | Xataka with Gemini In Xataka | Companies are turning their workers who know how to use AI into “stars”: the new labor gap

All the founders of OpenAI have become billionaires with ChatGPT. Everyone except Sam Altman, who has no shares

by

Sam Altman is the most recognizable face of the AI ​​industry in the world. He directs OpenAI, the company that created ChatGPT and is today valued at 852,000 million of dollars. However, a leaked document during the ongoing trial between Altman and Elon Musk Due to the change in status from an NGO to a for-profit entity, it has revealed who the true investors of OpenAI are and how much their participation in the company amounts to. In the box next to his name, only three letters appear: TBD, which in English means “to be determined.” The man who leads the biggest technological revolution in recent years does not own a single share of his own company. Sam Altman works for the love of art. OpenAI was born in 2015 as non-profit organization with an ambitious mission: to develop AI safely and for the good of humanity. That Altman did not take stock then made some sense since his role was presented as the neutral guardian, the leader whose decisions were not tainted by money. A noble mission, without a doubt. But that It is not the scenario in 2026. In 2019, OpenAI’s charity structure began to become too small to compete in the AI ​​race. OpenAI created a for-profit subsidiary under the so-called “capped-profit” model, in which investors they could make profits limited. That opened the door for capital and also for executives and co-founders to secure huge stakes in OpenAI. Altman’s name, paradoxically, remained blank. Those who did get rich, and a lot. As and how I collected Forbes, Greg Brockman, co-founder and former president of OpenAI, admitted during the trial that he owns a stake worth about $30 billion for which he paid nothing. Ilya Sutskever, former scientific director, has a participation between 30,000 and 35,000 million dollars. Figures very far from the annual compensation of $76,001 that its CEO receives, according to the tax form from OpenAI. The other major beneficiary is the Sound Ventures fund, linked to actor Ashton Kutcherinvested 30 million dollars in an early phase and that bet is now worth 1.3 billion, a return of 43 times the investment. In total, current and former employees control about $165 billion in company shares. The distribution among the greats. The block of corporate investors formed by Microsoft, SoftBank, Amazon and NVIDIA, together control 46.58% of OpenAI, with a stake valued at $396.9 billion against a combined investment of $122.7 billion. Microsoft leads that group with 26.79% of the company, a position valued at $228.3 billion built from an initial investment of 13,000 million. SoftBank occupies second place with 11.66% of OpenAI, valued at 99.3 billion compared to an initial payment of 64.6 billion, which represents a profitability of 1.5 times. amazon It has 4.66% of the company, valued at 39.7 billion dollars with an investment of 15 billion and a profitability of 2.6 times. At the top of the table is the OpenAI Foundation, the original non-profit entity, with 25.80% of the company and a stake valued at $219.8 billion which, having been formed with contributions without financial compensation, technically has an infinite return. Here may lie the key to the mystery of Altman’s retribution. A calculated move. The most widespread theory is that Altman and the board of directors, which he has firmly controlled since surviving the 2023 impeachment attempt, They are simply biding their time. Once the dispute with Musk concludes, the OpenAI board is likely to retroactively determine that Altman deserves participation commensurate with his responsibility. It is likely that, as is the case with other CEOs, This remuneration is linked to milestones like taking the company public with a valuation of more than a billion dollars. Perhaps this retribution will arise from that reserved fund now controlled by the OpenAI Foundation. Meanwhile, Altman is not exactly in trouble and your personal assets exceeds 2 billion dollars thanks to investments in companies that, curiously, are very well positioned to benefit from the growth of OpenAI. Without being a shareholder in his own company, he has built a personal business ecosystem that prospers directly thanks to his success. In Xataka | “The problem is Sam Altman”: more and more voices within the AI ​​industry are beginning to question the CEO of OpenAI Image | Flikr (TechCrunch)

OpenAI already knows which device will replace our smartphone in the age of AI. It will be another smartphone, according to Kuo

by

He doesn’t always get it right, but Ming-Chi-Kuo just made a particularly striking statement. According to your dataOpenAI is preparing its first “mobile AI agent”, a smartphone that will be quite different from the current ones not so much in form as in substance. If its predictions come true, we could be facing a device that will shake the pillars of the current mobile segment. Hello, “OpenAI phone”. Kuo states that mass production of this smartphone designed by OpenAI will begin in the first half of 2027. He also tells us that the SoC that will govern this device will be a customized version of the future MediaTek Dimensity 9600 manufactured with TSMC’s N2P process and that will theoretically arrive in the second half of the year. The mobile that wants to see the world. This chip will have some special features, such as an ISP (Integrated Signal Processor) with an HDR system that allows optimizing the visual perception of the world. It is logical: the mobile wants to become an integral part of our interaction with the world, and that visual capacity is critical. Two NPUs better than one. It will also have a dual NPU architecture to increase its AI computing capacity. It will theoretically integrate LPDDR6 memory and will have UFS 5.0 to avoid memory bottlenecks. If all goes well, Kuo says, between 2027 and 2028 30 million units will be distributed. Not anything else, but the plan seems incredibly ambitious. Paradigm shift. This type of device, Kuo points outwill doom the UI as we know it. The concept of navigating a patchwork of icons to perform independent tasks will be obsolete. The concept proposed by OpenAI understands that the user does not want to use an “application stack”, but rather achieve objectives through a centralized agent. This implies a radical redesign of the smartphone in which the screen stops being a menu of options and becomes a kind of mirror of what the user wants, of their “intentions.” We went from a manual interaction to a proactive inference, because the AI ​​is responsible for detecting what needs to be done to complete the action that the user needs. Without touching the screen. Task resolution rules over navigation. OpenAI being Apple. To achieve this OpenAI needs to control everything on this device, so similar to what happens with Apple and its iPhone. For an AI agent to function seamlessly, it needs access to sensors and device status in real time, something that current operating systems restrict by design. OpenAI wants to control both the hardware and the software to capture all the relevant information at all times. The technical barrier is not the AI ​​model, but that total control that also requires perfect management of memory and energy consumption. Apple, by the way, is in that same battle, although in a different way. The energy challenge. It seems logical to think that this device bases a good part of its capacity on AI models in the cloud, but also that it will have the ability to execute some tasks thanks to small local models. Hence having two NPUs that allow at least certain tasks to be executed on the mobile itself. That will be crucial precisely regarding energy consumptionbecause this AI that automates tasks by chaining them consumes much more computing than the usual interaction with an app today. App Store in danger of extinction. There is a particularly striking idea here. The app store economics faces existential disruption. The current model relies on friction: you need to open a specific app for each task, which justifies the 30% “tax” and the walled garden. If an AI agent can book a flight or order food by directly accessing the background APIs, the icon on the home screen disappears. The “app” stops being a destination and becomes an invisible tool. This not only threatens Apple’s revenue, but redefines mobile development towards an “API-first” ecosystem, where the graphical interface is irrelevant and competition is decided by agent efficiency, not UI design. Goodbye, privacy? And in this context, privacy could once again become the price of that “it’s so convenient to use a device like this” of these future mobiles. For an AI agent to be useful and function truly autonomously, it needs to know everything or almost everything about us. Our location, health, messages and of course the screen content at all times, among other things. The opacity of proprietary models will mean that we will never know what data is leaked to the cloud to “improve the service”, turning privacy into a variable controlled (once again) by the manufacturer. In Xataka | Microsoft has insisted on making Windows “agent.” His users have reminded him that they had not asked for it

Anthropic and OpenAI know that where AI is making money is in companies. They have found a way to squeeze that strategy

by

We end users no longer matter much to the AI ​​giants. These companies are confirming that income is currently in the professional world, and they are already making moves to conquer that segment. And if they have to do it company by company, so be it, because now OpenAI and Anthropic are a little less AI companies and a little more consulting. AI is more business than ever. Anthropic and OpenAI have understood that the real business of AI is not currently in individual $20 subscriptions, but in integrating their AI models into all types of corporations. Both companies have almost simultaneously launched alliances with other companies to provide consulting services. The objective is simple: to stop being external web tools to become the “operating system” of thousands of businesses through these exclusive sales channels. Anthropic on the one hand… The company led by Dario Amodei has formed a joint venture with Blackstone, Goldman Sachs and Hellman & Friedman valued at $1.5 billion. This new firm will act as a consultancy bringing Claude directly into the operating environments of mid-sized businesses, from mid-sized banks to local manufacturers to healthcare systems. These companies have committed to provide $300 million each for AI engineers to work closely with these clients to integrate custom solutions. …and OpenAI on the other. In turn, Sam Altman’s company has not been slow to replicate that initiative with the creation of the so-called The Development Company, an entity valued at about 10,000 million dollars. It is backed by funds such as TPG, Bain Capital and SoftBank. Theoretically, OpenAI has already raised $4 billion to accelerate the adoption of its AI models in more than 2,000 companies that are already part of those investors’ portfolios. The initiative is led by Brad Lightcap, until now COO of the company, and who wants to make the GPT family models an integral part of the operations of all types of companies. Engineers on the line of fire. To promote these strategies, both companies are adopting the so-called ‘Forward Deployed Engineer’ (FDE) model, a deployment system that was already popularized by Palantir and that consulting firms traditionally use. Instead of simply selling an API, Anthropic and OpenAI will send their engineers to work with doctors, financial analysts, or IT staff so that their AI models can be seamlessly integrated into those professionals’ real-world workflows. Going public as a goal. In recent months we seem to be experiencing a race against the clock towards the IPO in both cases. With absolutely stratospheric valuations (OpenAI 852 billionAnthropic hanging around 900,000 million), the pressure to justify these figures to the public market is immense. The integration of programming tools such as Claude Code has been a clear driver of recent growth, but the real gold mine is in the automation of processes in sectors such as health or finance. If you are joint ventures fail to scale quickly, the valuation bubble could deflate before those IPOs. Conflicts of interest. When a venture capital fund invests in a technology provider and simultaneously pressures its portfolio companies to adopt that same technology, competition ceases to exist. Many companies will not have much real choice based on product quality. What is reinforced here It is that “circular economy” in which innovation is not chosenbut is imposed by financial and business interests. The customer does not buy because he needs the tool, but because his own financial owner has a stake in whoever supplies that tool. But wouldn’t AI automate everything? The dependence on the FDE model is paradoxical. Theory tells us that software must be infinitely replicable at zero marginal cost. However, these alliances show that AI is still not smart enough to operate without direct human supervision. We need someone to teach us how to use it well, the companies say, and both OpenAI and Anthropic are going to take advantage of that need even if what we really have is luxury personalized consulting. For now, AI will be more part of the services offered by a consulting firm than a truly autonomous “plug and play” tool. New Job: Deployment Engineer. Now Anthropic and OpenAI will not only be AI companies: they will also be consultancies in need of manpower. That also serves as an example that although AI theoretically will eliminate jobswill also create new ones. Here we face a growing demand for “deployment engineers” —OpenAI already requests them—, professionals who are precisely in charge of adapting these AI models to the needs of companies that want to implement them in their daily lives. And the data, what. There is another fundamental problem: medium-sized companies will not have much capacity to manage their data sovereignty. For Claude or GPT to function properly in the business, they will need access to critical workflows, medical records, or sensitive financial data. And when one cedes that control to third parties, they remain vulnerable. Not only that: the security of this data is compromised because in order to process it, it must leave and be processed in the cloud of an external provider. The AI ​​models of these companies can also probably learn from these processes, although it is reasonable to think that Zero Data Retention policies will come into play (“No data retention”). Image | TechCrunch | Wikimedia Commons In Xataka | The White House wants to review new AI models before anyone uses them: first the Pentagon, then the rest of the world

OpenAI expects an 80% drop in its flagship revenue. The low-cost “ChatGPT Go” is your escape forward

by

OpenAI is in trouble. More than beforeeven. In The Information indicate that internal projections for subscribers in 2026 are worrying. The users of ChatGPT Plustheir $20 a month plan, will fall from 44 million in 2025 to just 9 million this year. That represents a drop of 80%, and they want to compensate for it with their affordable subscription. It’s not clear that plan can work. ChatGPT Go as a lifesaver. What OpenAI is going to lose with ChatGPT Plus according to these internal forecasts, they want to counteract with an extraordinary increase in subscriptions to ChatGPT Gothe ad-supported plan that costs between $5 and $8. The company’s objective is for this plan to go from having the current 3 million subscribers to 112 million, an increase of 3,600% in twelve months. A terrible quarter. While The Information showed these forecasts, in The Wall Street Journal they informed OpenAI does not have the accounts in this first quarter of 2026. The company has not achieved the expected income, and has not achieved the user acquisition figure that it had projected. OpenAI CFO Sarah Frier has warned that the company may not be able to pay for its future computing contracts if revenue doesn’t start growing immediately. The accounts do not come out. OpenAI has contracted close to $600 billion in spending on future data centers, an astronomical figure that was built with all the announcements that Sam Altman and the company made in 2025. The company expects to spend $25 billion but plans to enter $30,000, a narrow margin even if everything goes well. But according to WSJ it is not doing so, and Anthropic’s popularity has eroded its position in the market. They wanted to reach 1 billion weekly active users by the end of 2025 and they didn’t achieve it, and the decision to bet on ChatGPT Go seems like a desperate response to their revenue problem… and their IPO. No one has ever grown so much. ChatGPT Go’s growth goal poses a colossal challenge. Achieving 109 million paying subscribers in twelve months is unprecedented. It took Facebook four years to get 100 million free users, and although ChatGPT achieved the same thing in two months and set a prodigious precedent, for this to be repeated for a paid subscription even extending the time frame to 12 months would be unusual. But not even for those. Analyst Ed Zitron point Because even if OpenAI achieved 112 million subscribers at $5/month on average, it would earn $560 million per month. That figure is a far cry from the $880 million per month generated by the 44 million Plus subscribers at $20/month. The difference should be covered with advertisingbut that doesn’t seem to be going as well as they expected either. Until have activated pay per click adssomething that already caused the credibility of SEO to be greatly damaged. We go public, yes or no? According to WSJ, Sarah Friar and Sam Altman disagree about whether it is advisable to go public this year given this change in the situation. Altman wants to speed it up, but Friar doesn’t think the company is ready to meet the data reporting obligations that public companies have. The problems accumulate because the financing round closed in March made OpenAI’s valuation amounted to 852,000 million dollars. If investors had known the situation of OpenAI’s first quarter, perhaps they would not have entered that round, or they would not have done so in such a notable way. The challenge of charging $20 for AI. OpenAI’s forecast is worrying. That a company that managed to popularize generative AI can only get 9 million people around the world to pay $20 a month is disturbing and says a lot about the state of the market. On the one hand, maybe people just don’t see that $20 worth it, which is bad for the entire industry. But perhaps what people don’t see is that those 20 dollars are not worth it if they spend them on ChatGPT and they do on competitors like Claude. That is even more worrying. It is clear that there is a segment of users willing to pay such a price, but today that segment is smaller than the expectation created suggested. The Pro plan will remain a rarity. OpenAI also has the Pro plan for $200 per month, and expects its subscribers there to also double in 2026. However, that will still not be almost anecdotal because less than 1% of the total number of users—the truly intensive ones—will opt for this alternative. It is evident that this will not be the core of OpenAI’s business at the moment, and the company seems to be clear about this. They prefer to leave the middle segment in the background, have a small premium segment and bet on massive volume at a low price with advertising. We’ve seen this before… with Netflix. OpenAI’s strategy reminds us of the one Netflix launched with its advertising plan. Which many criticized when it was announced has become in a overwhelming success. The company has returned us to square one: we want to pay to see adssomething surprising but it works. And OpenAI seems to want to apply the same story. In Xataka | The surprise with the new GPT 5.5 from OpenAI is not that it is good: it is that Claude looks like GPT and GPT looks like Claude

We already know what happens to the GPU hourly price when OpenAI or Anthropic launch a new model: it doubles

by

This week, an analyst named Tomasz Tunguz published in X two revealing graphs. They show the evolution of what it costs AI startups to access cloud computing, and there is bad news. The cost of renting the NVIDIA B200 GPUs with Blackwell architecture has gone from $2.31 per hour in early March to $4.95 per hour this week. It is an increase of 114% in just six weeks and it has a clear cause: the arrival of new models from Anthropic and OpenAI. What the graphs show clearly. Those charts focus on the price index of Ornna cloud computing trading marketplace. The first of them covers the price of renting the B200 chips from the end of 2025 until today, and there are vertical lines showing each release of the latest models from OpenAI and Anthropic. The correlation is almost perfect: GPT-5 Codex, Claude 4.5, GPT-5.3 Codex, Claude Opus 4.7 and GPT-5.5 coincide with a jump in price indices. Every time these companies announce a new version of their frontier models, demand skyrockets, and so does the cost. If you want the best, pay (much more). The second graph shows the price difference between renting the previous generation of chips, H200 with Hopper architecture, and the new B200. The historical average of that “spread” is $1.06, but now it stands at $2.09, practically double. That means buyers—startups and AI companies—are paying a record premium for the extra memory and superior computing power of Blackwell architecture chips. Accessing the latest of the latest was already expensive. Now it is even more so. This also makes the H200 in a second class option for the most demanding models of 2026. Action and reaction. There is overwhelming logic here. When OpenAI or Anthropic release a new model, there is an explosion in inference. Developers and companies want to test them as soon as possible and integrate these models into their products (or compete with them). To do this, they need computing quickly, and a simultaneous demand is caused that unbalances the available inventory in the market for renting AI chips by the hour. The problem is that the supply of B200 does not grow at the same rate. Some companies have wanted to anticipate, and we have the perfect example in Google. He has bought all the B200s he can, and that has made these GPUs around now the 500,000 dollars on the secondary market according to analyst Jack Minor. The irony of efficiency. The curious thing is that the more efficient these chips are – and the B200s are – the more companies want to rent them at the same time to take advantage of those efficiency advantages that should lead to cost savings. What actually happens is that the scarcity of these advanced chips cancels out any theoretical savings. Long term contracts. Startups and companies that think in the short term are especially harmed in this area, because they face price jumps that are increasingly difficult to assume. Companies that signed computer rental contracts at the price then can now operate at less than half the cost of their competitors. Thinking in the medium or long term seems reasonable, although once again those who win are the hyperscalers and those companies that have managed to get hold of many B200s. And who wins even more is of course NVIDIA, which cannot cope. Few alternatives. In other markets such as energy or metals there is usually room for maneuver, Tunguz points out, but the same is not happening at the moment in the AI ​​segment. In the oil market, for example, if the price rises 114% in six weeks, companies can buy futures, options or fixed-price supply contracts to protect their margins. In cloud computing rental, those options are much more limited. And the result is a much more volatile segment. This will go further. We are probably facing a peak in demand that will be followed by a correction: the new batch of B200 chips that arrive in the second half of 2026 are expected to cause a drop in current prices. However, that $4.95 is now the new floor, not a peak, because demand for AI computing will continue to grow faster than TSMC’s production capacity. In the absence of the supply of AI chips growing significantly – and there are certainly movements that are trying to achieve this, such as those of Google with its TPUsAmazon with its Trainium or Huawei with its Ascend—, the problem will still be there. In Xataka | Europe is taking its technological independence so seriously that it is aiming for the most ambitious goal: NVIDIA

AI already knew how to create images. OpenAI says it has found the missing piece with the new ChatGPT Images 2.0

by

Over the last few years we have seen image generators become increasingly more spectacular, faster and also more popular. The problem is that a striking image is not always useful to work with. It is one thing to ask for an astronaut cat and quite another to obtain a usable marketing poster, a coherent vignette or a graphic that respects what we have asked for. That’s where OpenAI now wants to move the conversation with its new model: not so much towards the pretty image, but towards the useful image. The answer. What OpenAI proposes goes in that direction. The company led by Sam Altman He maintains that his new model is not only created to generate attractive images, but to solve visual assignments with more intention and less trial and error. In the presentation he went so far as to state that “images are a language, not decoration”, a fairly clear way of summarizing where he wants to take the product in a present with quite a bit of competition. The thesis is that: that asking for an image in ChatGPT It’s less like launching a creative prompt and more like commissioning a piece that we can actually use. The missing piece. If the firm wants us to talk about something more than showy images, it had to improve exactly the points where these models usually fail. Here they promise important changes on three very specific fronts: following complex instructions more precisely, better organizing elements within the image and reproducing dense text with greater reliability. In other words, we are not only looking for more beautiful results, but also less ambiguous and more controllable ones. Think before you draw. One of the novelties that OpenAI tries to highlight most strongly is that this is its first image model with reasoning capabilities. Translated into practical terms, the company maintains that, when a model with “thinking” is chosen within ChatGPT, the system can take more time, structure the task better, rely on the web to search for updated information and review its own results before delivering the image. And we have tried it, asking for the image of two people walking along Gran Vía, in Madrid, near Cines Callao, and some notes on activities to do in Spain during May. These are the images that we can see in the cover image. The keys. OpenAI talks about game prototyping, storyboards, marketing creatives, comics, social graphics and other materials where both content and form matter. To sustain that ambition, the company says it has improved on two delicate fronts: the handling of non-Latin text, with advances especially in Japanese, Korean, Chinese, Hindi and Bengali, and the more faithful reproduction of very marked visual styles. It also expands the possible formats, with proportions of up to 3:1 and 1:3, resolution of up to 2K and, in certain modes, the possibility of generating up to ten images within the same request with continuity between characters and objects. The competitive context. This announcement also cannot be read as if OpenAI had suddenly discovered a new market. Midjourney has already become a clear reference for works with a strong artistic charge, Nano Banana has attracted attention for its conversational editing capabilities and FLUX 2 has become strong in photorealism. With that board in front, the company seems to be looking for another angle. Rather than contesting each terrain separately, it tries to present ChatGPT as an environment where the image is not generated in isolation, but as part of a broader flow, something that on paper can be attractive if it really delivers what it promises. It’s already starting to unfold: One of the keys to the announcement is that OpenAI ensures that the model does not remain in the showcase phase, but is beginning to reach a product. The company places its deployment in ChatGPT for all users, including Free and Go, and associates the most advanced results with Plus and Pro, as also reported by Engadget. Additionally, it takes you to the API and Codex, a sign that they don’t want to limit it to casual use within the chat. If your strategy involves turning the image into another work tool, it made sense for the deployment to start precisely there. Images | Xataka with ChatGPT Images 2.0 | OpenAI In Xataka | Amazon wants to win the AI ​​race at any price. That is why it has invested both in Anthropic and OpenAI

chatbot is not working and OpenAI says it is investigating an issue

by

If you are a user of ChatGPT and this afternoon you wanted to ask the chatbot something, you’ve probably been left without an answer – it’s time to use your brain again. It is that the famous service powered by artificial intelligence (IA) has been giving errors for several minutes. OpenAI, for its part, has launched an investigation to understand the origins of the problem. The outage entered the scene around 4:00 p.m., preventing users from around the world from using ChatGPT normally. As we can see in the screenshot, the chatbot refused to respond, offering error messages such as ‘Hmm… something seems to have gone wrong’, which in Spanish means ‘Mmm… something seems to have gone wrong’. In development. Images | Solen Feyissa In Xataka | What is Cloudflare, how it works and why a crash or block causes half the Internet to fail

The US has appointed executives of Meta, Palantir and OpenAI as lieutenant colonels. We have many questions

by

On June 13, 2025, four executives from some of the world’s largest technology companies donned the uniform of the United States Army at Myer-Henderson Barracks, a ten-minute drive from the Pentagon. After taking the oath, They were appointed lieutenant colonels of the Reserve. The appointment was controversial, but it was made on the occasion of the launch of Detachment 201, a very special army body dedicated exclusively to military innovation. Technological military with a wink. The four new reserve lieutenant colonels are Shyam Sankar, CTO of Palantir, Andrew Bosworth, CTO of Meta, Kevin Weil, CPO of OpenAI and Bob McGrew, advisor to Thinking Machines Labs (Mira Murati’s startup) and former head of research at OpenAI. The name of Detachment 201 is a wink to Silicon Valley, because an HTTP 201 status code on the web means that a resource was successfully created. All four will continue in their current positions while serving as reservists. Sankar’s thesis. Palantir’s CTO has already become a reference in the discourse on how to apply technology to military institutions after publishing on its website 18theses.com the document “Defense reform”. In it he talked about how “warriors fight with weapons and with git.” He criticized the Department of Defense (DoD) for treating technology as “expensive and unaffordable,” and proposed using AI to make military assets work more efficiently and quickly. The germ. The project was conceived by Brynt Oameter, who was responsible for talent management at the Pentagon. His idea was to attract technology experts so that they could take up positions in the Army when necessary. He met Sankar at a conference in early 2024 and began discussing the idea, which ended up crystallizing into a project that Donald Trump promoted. Finger designations. A curiosity: among that group of chosen ones there were no Anthropic executives even though the company was the one that ended up being the chosen one in July 2025 to integrate its AI model, Claude, into Pentagon systems. Then, how do we know, things changed. On Wired they explain how Sankar was the one who volunteered to be part of the project, but also recommended the three people who would end up forming that group with him. What will these four managers do in the Army?. The official mission of these experts is to integrate specialized knowledge in AI, software and data analysis into the Pentagon’s strategy. parameter gave an example: The commander of the Indo-Pacific region is evaluating threats in the Far East for the next ten years and has asked Detachment 201 to explain how AI can affect security in that context. These new lieutenant colonels can also operate more tactically, advising on how soldiers can use the new tools at their disposal. Or what is the same: they will act as consultants to the US Army, but in uniform and having taken the oath, something important because the relationship with the soldiers changes. The inevitable conflict of interest. The Army affirms that there is no conflict of interest because the members of Detachment 201 will not have a vote in the contracts signed with the private sector. The chronology of events tells us otherwise, however: A month before Bosworth took office, Meta announced an agreement with Anduril to develop military augmented reality products. A few months before OpenAI announced an alliance with Anduril in air defense systems. Palantir, Sankar’s company, signed a contract with the Army worth 480 million dollars in December 2024. That doesn’t prove anything, but suspicions are inevitable, because even if they don’t have a vote, they will be able to obtain internal knowledge and data that inevitably benefits their employer companies. But weren’t there going to be limits on AI in the army? Another of the thorny questions that arise from this Detachment 201 is how the recommendations of these experts will be applied on the battlefield. OpenAI theoretically has policies prohibiting its AI models from causing harm or developing military weaponry. However, the explicit mission of this body is to make the US Army be “more lethal”. That contradicts OpenAI statementswhich after allying itself with the Pentagon recently stressed again and again that its models would be used within limits… which is exactly why the Pentagon ended up wanting to turn Anthropic into a pariah company. Two weeks to get the rank. A conventional lieutenant colonel reaches that rank after between fifteen and twenty years of active military career. The members of Detachment 201 received that same rank after two weeks of partially online training that included physical conditioning, shooting as a diagnosis and basic notions of military protocol such as the rank structure and the use of the uniform. They did not complete basic training and have the flexibility to fulfill part of their 120 annual hours of service from home, something not offered to other reservists. All of this has generated reviews within the Army and also comments of all kinds on social networks. Image | DVIDS In Xataka | Anthropic and OpenAI have developed AI. The US Pentagon is showing you who really owns it

OpenAI swore that ads on ChatGPT were its “last resort.” Now they are your survival plan

by

a couple of years ago Sam Altman said that placing ads on ChatGPT was “the last resort for our business model.” Well then, ChatGPT ads are here and OpenAI is sure that it will be the business of the century, one that will generate a whopping $100 billion. what has happened. He leaked it Axios; During a presentation with investors, OpenAI has confirmed its forecasts for the newly released advertising model in ChatGPT. During 2026 they expect to generate 2.5 billion dollars and this will increase in the coming years until reaching 100 billion in 2030. This is the progression they project: 2026: 2.5 billion 2027: 11,000 million 2028: 25,000 million 2029: 53,000 million Why it is important. Advertising has gone from being the last resort of its business model to directly being its business model. OpenAI is losing money at an unsustainable rate and has been making profound changes to be more profitable, such as focus more on enterprise customers, but it may be too late. Advertising is your way to profitability. In other words, your survival depends on this going well. butterfly effect. If it works for them and they achieve their goal, it can change the rules of online advertising. 100,000 million is many millions, enough for Google and Meta’s business to end up being affected. Furthermore, advertising within a chatbot like ChatGPT can be much more profitable because the user says in a much more direct and detailed way what they are looking for. On the other hand, advertising on Instagram or Google Ads requires work to collect data to guess the user’s tastes. If it doesn’t work for them, the outlook looks bad for the technology sector. We talk about the most valuable private company in the world and Its possible bankruptcy can cause a domino effect that freezes investments and punctures the expectations placed on AI. Users. To achieve these numbers, OpenAI estimates that it needs its weekly user base to reach 2.75 billion by 2030. Right now ChatGPT has 900 million weekly active usersthat is, they have to triple them in four years. We talked about ChatGPT having to be at the level of WhatsApp or YouTube. There is already 6 billion people with internet accessAs far as there are users, the question is whether it is feasible for OpenAI to attract almost half of them. The mass adoption of AI is already in a more mature phase and, although it is the most used, ChatGPT is no longer the pretty girl; Now it coexists with equally capable competitors and most importantly: The image of the company has been eroding. The double edge of advertising. Advertising can be tremendously lucrative for OpenAI, but it puts user trust at risk, and that is just what they need to fulfill their plans. We have normalized seeing ads everywhere, but having them appear in a conversation with a chatbot threatens to erode their main promise: to be assistants that respond solely to the user’s interest, and not to the commercial priorities of those who pay to advertise. Two things can happen here: that the rest of the AI ​​companies jump on the bandwagon and we normalize that the free versions have ads (the ideal scenario for OpenAI), or that OpenAI is left alone and people end up going to other ad-free chatbots. Anthropic said it would not advertisewe will see if in a few years they continue to maintain it. Image | Xataka In Xataka | Before, advertising was to monetize. Now it is to punish you and YouTube has taken it to the extreme

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.