OpenAI Faces $140 Billion In Losses Before It Turns $1 In Profit

Earlier today we presented the musings by IBM CEO Arvind Krishna, who predicted that at today's cost/benefit calculus, investing trillions of dollar in data center capex will never be a profitable venture. 

But some don't have to wait years for this revelation: they have already found this out, and as Deutsche Bank's Jim Reid shows in his chart of the day, "it’s been eye-opening to read of the predicted losses that OpenAI will likely experience in the next few years."

Based on a Wall Street Journal report, which cited company projections provided to investors over the summer, the DB strategist notes that OpenAI forecasts revenue of $345 billion between 2024 and 2029. Assuming this is all cash, we then calculate $488bn of spending, mainly to pay for access to compute, to arrive at their projected cumulative free cash flow of $143bn. And that was before the most recent announcements of $1.4 trillion in data center commitments.

Previously, HSBC said the cash burn could exceed $210bn by 2030.

So in the context of OpenAi's funding black hole, Reid thought it would be interesting to look at the largest cumulative losses in history from a young company or a start-up before they turned in a profit. So he asked ChatGPT to give a table of these companies, detailing the total losses and over which years. He then used this to create today’s CoTD, and added in OpenAI’s expected cash burn and also included its rival Anthropic, also using data from the Journal. 

ChatGPT also pointed out that some companies had reported larger annual losses, citing AOL Time Warner’s $99 billion loss in 2002 and a similar size loss for AIG in 2008. Meanwhile, Fannie Mae and Freddie Mac lost $77bn and $59bn respectively within only three quarters when the GFC struck. However, these were well established companies with long track records of profitability before huge troubles hit.

OpenAI is at a tricky point in its development three years after the launch of ChatGPT made it the poster child for the boom. Another member of Reid's team at DB, Adrian Cox, published a fascinating report yesterday (available to pro subs) unpicking the challenges it faces, including subscriptions, substitutes and substantial costs. "Happy Birthday, ChatGPT?” includes exclusive transactions data from dbDataInsights (dbDIG) showing how subscriptions in major European markets have more or less flatlined since May. When Adrian first shared this proprietary OpenAI subscription data a month or so ago, it generated a lot of interest. The updated figures now suggest that the earlier data was unlikely to be an anomaly.

Of course, OpenAI may continue to attract significant funding and could ultimately develop products that generate substantial profits and revolutionise the world. But at present, no start-up in history has operated with expected losses on anything approaching this scale. We are firmly in uncharted territory.

More in the full DB note available to pro subs.