OpenAI has concluded a $122 billion (£92bn) funding round at a post-money valuation of $852bn, as the fast-growing but unprofitable start-up faces increasingly complex challenges in maintaining its lead in the AI boom.

The firm previously announced $110bn of the funding in February, including an investment of $50bn from Amazon that involves deploying enterprise tools using Amazon Web Services infrastructure – a plan reportedly being challenged behind the scenes by Microsoft.

Some $35bn of Amazon’s commitment is contingent on OpenAI holding an initial public offering or achieving the ill-defined milestone of artificial general intelligence.

Nvidia and SoftBank both contributed $30bn, with additional funding from Andreessen Horowitz, Abu Dhabi’s MGX, D.E. Shaw Ventures, TPG and T. Rowe Price and others.

The start-up also raised more than $3bn from individual investors through bank channels, the first time it has done so.

It said it would be included in exchange-traded funds managed by Ark Invest as a means of offering further exposure to individual investors.

OpenAI has previously committed to some $1.4tn of infrastructure investments, including spending on data centres and AI chips, but has remained vague on how the spending would be financed.

To help obtain the necessary funding, OpenAI and rival Anthropic have entered into deals with their own suppliers of cloud infrastructure and AI chips, including Amazon and Nvidia, raising questions around the sustainability of today’s rapid AI-sector expansion.

Last year, OpenAI and Nvidia announced a multi-year deal worth $100bn, much of which was to be re-invested in buying Nvidia chips, but the companies abandoned this plan in favour of Nvidia’s most recent investment in the company.

Anthropic recently shut down its Sora video-generation tool, a surprise move that The Wall Street Journal reported was largely due to the low usage and high cost of the service.