C3.ai: not every AI company is a growth stock

If artificial intelligence is the hottest sector of the year, why is self-described AI software company C3.ai finding it difficult to grow sales? The California company’s share price is up 229 per cent this year, despite a challenge from short sellers. Bagging the ticker symbol “AI” did not hurt. Yet revenue growth in the past fiscal year was less than 6 per cent.

C3’s problem may be that it is trading in the wrong kind of AI. Unlike OpenAI’s large language models and generative AI chatbots, C3 sells software that analyses data. It has launched a generative AI service, but this is not a core product. Revenue growth this year is forecast to be about 14 per cent at best.

When it listed, C3 attempted to drill the idea of enterprise AI into prospective investors and customers, peppering San Francisco with billboards. This is not, however, what it was made for. Founder Thomas Siebel wanted to offer software services to companies addressing carbon emissions — the “C” in C3 is for carbon. The company has gone through a series of rebrandings since then, including a brief spell as C3.IOT. In 2019, it named itself C3.ai.

Pivots are lauded in Silicon Valley. But the company spends a lot to bring in new customers, something that does not tally with heightened global AI interest. In the 12 months to April 30, it spent $183mn on sales and marketing, equal to 70 per cent of revenue. Growth is uneven. A small number of customers account for a large proportion of sales. Three made up 44 per cent of revenue when the company listed in 2020. In the past quarter, just one company, Baker Hughes, accounted for 45 per cent. The joint venture will end in April 2025.

Now that it is moving to a model that charges by consumption, revenue could become less predictable. There is no date in the diary for sustainable profitability based on generally accepted accounting principles. C3 illustrates the gap between AI excitement and sustainable, AI-driven income.