Post

Based on analysis of years of revenues, losses and funding, since 2023, OpenAI has taken in $28.6bn in cash and had $13.7bn in losses. However, reports say OpenAI ended 1H 2025 with $9.6bn in cash - suggesting OpenAI's losses are billions higher than reported. www.wheresyoured.at/where-is-ope...

OpenAI is has now signed over a trillion dollars of cloud deals it can't afford. Based on these costs, OpenAI will spend $11 billion a month by 2027. This is more than the operating expenses of Meta, a profitable company with 78k employees. www.wheresyoured.at/where-is-ope...

In an analysis of OpenAI's costs, revenues and funding from 2023 to 1H 2025, it raised $28.6bn in cash ($10bn rev included), and had $13.7bn in cash losses. Reports say OpenAI ended 1H 2025 with $9.6bn. It should have $14.9bn in cash. There's a $4.1bn+ hole. www.wheresyoured.at/where-is-ope...

Based on my analysis, OpenAI burned $19 billion of their $28.6 billion in cash from 2023 through 1H2025, meaning there's at least $4.1 billion in cash unaccounted for, as my analysis includes *ALL* reported costs that OpenAI has incurred. Where is the money? www.wheresyoured.at/where-is-ope...

I don't understand, but I want to. What does inference mean, in this case? Is there a "for dummies" version of this?

A simplified version. You *train* a model to do something. Example, differentiate dogs from cats in photos. This might take 1000s of photos. You then use a model on a new photo. “Is it a cat or dog in this photo?” That is “making an inference.” Inferences are claimed cheaper than training. 1/3

This is the DVD model. It is cheaper to playback a DVD (compute-wise) than record it. Good, because you record once and play many times. If LLMs train expensively but do inference cheaply, LLM firms can say, “We did the expensive part already, now on to the cheap (profitable) stuff.” 2/3

OpenAI CFO Sarah Friar claims that OpenAI could have "healthy gross margins" if it "wasn't seeking to invest so aggressively." I believe the truth is that OpenAI's on path to getting more out of their models is more inference - which will make margins *worse.* www.wheresyoured.at/where-is-ope...

Reports say that OpenAI will need to raise another $70bn in 2027. Based on my calculations, I believe OpenAI needs to raise $100bn in 2026 just to handle its rate of spending, and potentially more based on $1.4tr+ in deals it has agreed to. What a mess. www.wheresyoured.at/where-is-ope...

As a note: that $4.1 billion in missing cash isn't covered by data centers or acquisitions, either. Oracle/Crusoe building Stargate, Microsoft builds the rest of the capacity (or CoreWeave does), and all acquisitions have been in stock. Very strange stuff. bsky.app/profile/edzi...