🫧
🔹 Hedge funds and large investors are overloaded with bets on AI infrastructure. The level of capital concentration is close to extreme values. The problem is that if the market starts to cool sharply and investors want to exit, it will be almost impossible to do so. There will be many more willing to sell than those ready to buy. There simply won't be enough money and liquidity to "digest" such a volume of exits.
🔹 Competition is shifting from model quality to the cost of their operation. With comparable capabilities, the decisive factor becomes the cost per query, not technological advantage.
🔹 Google is in a structurally stronger position. The company uses its own TPUs, owns data centers, and effectively pays only for electricity and depreciation.
🔹 OpenAI remains vulnerable in terms of costs. The company spends about $3.3 to earn $1 in revenue. The high margin of Nvidia, the markup of Azure, and its own operating expenses form a model where scaling worsens the economy.
🔹 Funding is closed-loop. Investors put money into OpenAI, OpenAI directs funds to purchase GPUs from Nvidia and Microsoft infrastructure, and capital returns through suppliers to the same financial centers. This structure increases systemic fragility.
🔹 Risks are taking on a systemic scale. The connection between OpenAI, Microsoft, Nvidia, and large funds creates a circular shoulder effect without a natural buffer, increasing the likelihood of a chain reaction during a sharp revaluation, comparable in logic to the 2008 crisis.
🔹 Legal risks are increasing. The lawsuit between Elon Musk and OpenAI has reached the jury trial stage, and the amount of potential claims in the case is comparable to the money that the company has in its accounts. That is, in the worst-case scenario, this process could become not just a legal issue for OpenAI but a serious financial blow.
🔹 ChatGPT traffic in the US is declining, Gemini is rapidly gaining an audience, key figures are distancing themselves, and alarming signals are being heard within OpenAI.🪰
