lowRISC® was created with a mission to help make open-source silicon designs a reality, commercially relevant, and widely adopted throughout the industry.
This article explores why we think that open silicon is a good thing, what a future in which open silicon is commonplace looks like, and how far we are from such a future.
Why open silicon
We believe that the same powerful reasons that have made open-source software commonplace also apply to hardware.
Open silicon IP components and complete chip designs can bring huge benefits to the semiconductor industry:
- Cost reduction: by eliminating expensive IP licensing fees and royalties, whilst enabling engineering collaboration and reuse of complex infrastructure (e.g. continuous integration and regression testing) on complex SoC designs
- Transparency throughout the value chain: enabling the company responsible for the final product to understand and build trust in the design of each component
- Flexibility: to adapt the IP to be exactly the way you need it, and achieve the best compromise between features, performance, power, area and security
- Quality and security: having many eyes on the design to spot defects and vulnerabilities, and raising the bar for commercial alternatives
- Faster innovation: by accelerating evaluations, and getting access to complete chip designs rather than only IP blocks
- Lower risk and greater choice: by not being locked to a single IP vendor, silicon vendor, engineering contractor or foundry
- Education: training of new engineers in production-quality IP, processes and systems
Current state of the semiconductor industry
The semiconductor industry has already made three significant steps, which lead us to believe that we are on the threshold of the golden age of open silicon.
Broad adoption of the IP business model
All chip designs today are modular, and the vast majority adopt third-party IP from IP vendors like Arm and Synopsys, and foundries like TSMC. This IP reuse has brought to the industry a major reduction in development cost and time and increase in quality, which are benefits typically associated with open-source.
IP vendors often make their design files available to their licensees, which has built over time a high degree of trust in the vendors themselves and their products.
It is only a small step for silicon vendors to start using open-source IP instead of that from commercial vendors. It is “only” a matter of building the required trust in the quality of the design. Thanks to modularity, silicon vendors can adopt open-source IP block by block, managing their risk and gradually building trust.
By using open-source IP, silicon vendors will benefit from further cost reductions, transparency with their own customers, flexibility in the use of the IP, along with increased quality and speed of innovation.
Introduction of the RISC-V CPU architecture
Every SoC needs a CPU but the CPU is not the key differentiating factor for most chip designs. Access to commercial quality open-source CPU designs can significantly lower the barrier to entry for new silicon vendors, thereby fostering innovation.
Open-source CPUs require an open-source architecture. The cost of creating a successful ecosystem around a CPU architecture is huge, but this work has already been done.
The RISC-V architecture was introduced in 2010 and is now mature in the embedded domain, in terms of number and variety of CPU implementations, tools support, stability and governance. RISC-V processors have been included in billions of chips.
First commercial adoption of open-source designs
The Ibex® CPU was released by lowRISC in 2019 after extending and verifying the original zero-riscy core. Since then, it has been adopted in multiple SoC designs.
In 2024 open silicon adoption took a big step, when discrete root of trust devices from Nuvoton based on OpenTitan® reached commercial availability, and in 2025 we saw the first release of the Caliptra secure enclave design which also includes elements from OpenTitan.
These successes show that both large corporations and small vendors are starting to use open silicon in increasing numbers, and many of the old barriers to adoption have been knocked down.
Barriers to adoption of open silicon
Unfortunately, hardware introduces some tough challenges that have made the adoption of open-source practices slower than in software.
The most important ones are that the cost of going to production is very high, and that the cost of updating hardware in the field is prohibitive, which forces silicon vendors to be very cautious with quality and IP risk.
It is not feasible for silicon vendors to simply take some RTL from GitHub and put it in their SoCs. They need to get strong reassurances of:
- The legal processes through which the IP was contributed, to minimize the risk of infringing copyright or patents
- The verification the IP has gone through to demonstrate the quality of the design
In practice, it requires commercially minded organisations to run open silicon projects as a product intended for real deployment.
As recent history suggests, the open silicon projects which get commercially adopted are the ones that implement the right processes and have serious companies onboard to take them to market. We explore this in our Silicon Commons® article.
A vision for open silicon
At lowRISC we believe that open silicon IP and complete chip designs will become commonplace in similar timescales to open-source software and that, as it happens today with software, most SoCs in the future will incorporate some open silicon IP within their design.
Designers looking to get to market quickly with the best possible, most competitive products, will leverage high-quality IP from the open-source domain, avoiding reinventing the wheel. They will concentrate their scarce design, validation and firmware resources on novel unique selling points that are truly innovative and value-additive.
Open silicon will not spell the end for commercial IP and silicon vendors, just as open-source software did not spell the end of software vendors. Rather, it will raise the quality bar for everyone, forcing commercial vendors to up their game and create valuable differentiation and no doubt incorporating foundational open silicon designs in the process.
Future graduates will learn and use open-source hardware IP in their engineering degrees and will be better trained on the development processes used by IP and silicon vendors. This will further enhance the productivity of the semiconductor industry as a whole.
With designs like Ibex, Caliptra and OpenTitan proving that commercial-grade open silicon can be built and go into mainstream sockets and SoC designs, the existence proof and initial toolkit of high-quality IP blocks now exists.