The technology world is focussed on AI right now, but the more we read about the subject the more we find ourselves asking how are we going to get to that bright, shining, GPT-powered robot future? All of that will need to be built on foundations made of some very solid hardware.

We recently spent time recently trolling through Industry Research. According to the analysts at IDC, over the next five years 60% of revenue in “Technology” will come from Hardware, with only 40% coming from software. But all of that future technology will need to be built on products being developed over the past few years. And by contrast, over the past 5 years 90% of Venture investments went to software, with only 10% going to hardware. And of course, the number of software investors is several orders of magnitude larger than the number of hardware investors. There are 1,000 institutional venture funds in the US, only 10 of these can credibly claim to be hardware investors. This is a massive mismatch that needs to be addressed.

To be fair, much of the investment for hardware will come from large companies, not venture investors. But this only goes so far. The big hardware companies either have margins too tight to allow for serious R&D (e.g. the PC makers) or they compete from duopoly or near-monopoly market positions (e.g. many semis companies). Monopolists do not have a great track record when it comes to innovation. We cannot rely on these companies to make the investments needed to fuel the future.

The US government would very much like to encourage investment in hardware, but these ambitions can only go so far. The US Chips Act provided $50 billion over 5 years, which is a large sum but has to cover so many bases that it will be insufficient to plug the gap.

By contrast, the other major source of hardware funding is incredibly eager to invest huge amounts – of course, that source is the government of China. We do not want to dive into the rabbit hole of geopolitics and the Thuycdides Trap, but if US and European entities do not invest in hardware, China will end up being the source of much of this investment and thus likely capture all that value.

US investors need to take a long, hard look at upgrading their capabilities. And all the pensions, endowments and sovereign wealth funds who provide Limited Partner funds to those VCs, need to dig a bit deeper into their portfolios. Most will find that at least 90% of their “venture” allocations are invested in software, creating a major misallocation. There is a path out of this mismatch, but the venture community needs to retool to capture the opportunity.