AI opens up chip design, threatening the grip of a few big players

Chip design is one of the most closed and expensive disciplines in the tech industry. Electronic design automation (EDA) tools have traditionally cost tens of millions of dollars a year, and the processes themselves demanded deep expertise available only to a few large teams at companies such as Intel, Qualcomm, Apple or TSMC. AI is starting to change that picture.

According to Wired, artificial intelligence is already being used to simplify the design of silicon circuits and to optimize software for different processor architectures. This is not just a speedup of existing processes. It lowers the entry barrier itself in a market that has historically been closed to everyone except tech giants and government labs.

Several startups see this as the basis for a full-blown revolution in chipmaking. Their logic is simple: if AI can take over the routine and technically complex parts of EDA workflows, such as circuit verification, component placement, signal routing and optimization for a specific manufacturing node, then small teams could design specialized chips without years of investment in expertise and costly licenses.

At the same time, AI is helping with another problem: optimizing existing software for heterogeneous hardware. As the chip market fragments, with ARM, RISC-V and custom accelerators from Google, Amazon and Microsoft, developers find it harder to squeeze maximum performance out of each architecture. AI tools promise to adapt code to specific hardware automatically, relieving developers of manual optimization.

If those promises pan out, the consequences for the industry will be significant. Today a few companies control the EDA tools market, and developing specialized chips remains the privilege of corporations with multibillion-dollar budgets. Democratized access would mean that university labs, tech startups and even individual engineers could build custom hardware, something that used to take years-long cycles and hundreds of millions of dollars.

The shift fits a broader trend: AI is steadily lowering barriers in the most technically demanding fields, from biology and materials science to software engineering. The semiconductor industry turns out to be next on the list.

How far the democratization of silicon goes will become clear over the next decade.