OpenAI, the company behind ChatGPT, could go bankrupt within the next 18 months claims a new report.

According to an in-depth breakdown by Analytics India Magazine, OpenAI spends around $700,000 per day to run ChatGPT. Its other products such as DALL·E 2 are not factored into that number.

The report notes that while it is being bankrolled by partners with deep pockets – chief among them being Microsoft which has invested $10 billion – the company is deeply unprofitable. In May, the company reported that its annual losses doubled to around $540 million.

While company boss Sam Altman has reportedly suggested that OpenAI may try to raise up to $100 billion in the coming years, a route to IPO has not been forthcoming. A seperate Investopedia report noted that it takes at least 10 years of operation and $100 million in revenue for an IPO to be successful, and based on its current operating costs OpenAI would burn through its cash reserves well before that.

And while it emerged as an early frontrunner in generative AI, OpenAI could potentially be eclipsed by the established big tech names like Google and Meta which are pouring billions into their own large language models (LLMs).

Fears have also been compounded by a steady decline in ChatGPT site visitors over the past two months – from a peak of 1.9 billion monthly website visits in May to 1.5 billion in July. The report speculates that this could be due to school holidays, or that, since the release of the ChatGPT API for users, people are starting to build their own bots instead of using the core ChatGPT product.

Global GPU shortages are also a factor. With OpenAI now looking to train the upcoming GPT-5, resources have been turned towards that and the output quality of the existing ChatGPT product has declined.

Recent news from the business world does not make for pretty reading for OpenAI either. According to a study conducted by BlackBerry of IT decision-makers in Australia, Japan, France, Germany, Canada, the Netherlands, US, and UK, 75 per cent of respondents are implementing or considering plans to prohibit generative AI applications in their workplace. Some 61 per cent said their measures would be permanently implemented or long-term.

Share Story: