PureLiFi never lacked believers. For fourteen years the Edinburgh spin-out raised more than £35 million from investors, funds and banks to develop its pitch that light, not radio waves, could carry the next generation of wireless data. On 31 August the company’s directors ran out of options, appointing administrators after failing to secure a further round. All 42 staff at its Leith head office have been made redundant, as reported by the Edinburgh Evening News, The Herald and BusinessCloud.
The technology behind the collapse is genuinely interesting. LiFi transmits data by varying the intensity of LED light at speeds the eye cannot see, offering a bandwidth highway beyond the crowded radio spectrum that carries wi-fi, 4G and 5G, and one that cannot leak through a wall, which appeals to security-conscious buyers. The company was founded in 2012 by Professor Harald Haas, widely billed as the father of LiFi, and Dr Mostafa Afgani, out of the University of Edinburgh.
What killed it, according to the administrators, was a strategy pivot. Kenny Craig and Kevin Mapstone of BTG, the joint administrators, explained that pureLiFi had shifted from licensing its globally patented technology to third-party manufacturers into building its own hardware. That move “was more costly than initially anticipated”, and by the end of the second quarter of 2026 the business had simply run out of cash flow. Thomas McKay, BTG’s managing partner for Scotland and Northern Ireland, put it plainly: the firm “ran out of money before it could cross into profitability”.
The failure stings because public money was involved. The state-owned Scottish National Investment Bank backed the company and now faces losses it has already provisioned in its annual accounts, telling the press that “market challenges proved too great” despite working with the firm and other stakeholders to review options.
Deep-tech’s structural problem is on full display. Fifteen years, dozens of patents and a real product still do not answer the question a hardware business has to survive daily: can you manufacture at scale, at price, faster than the balance sheet drains? Investors fund the science. Almost nobody funds the decade of unglamorous manufacturing it takes to commercialise it. The light-based internet is not dead, but Wednesday’s news is a reminder that in deep tech, being right about the physics is the easy part.