Honda Motor Co., Ltd. and U.S. analog‑AI company Mythic announced a joint development agreement to create an automotive‑grade system‑on‑chip (SoC) that will deliver more than 100,000 TOPS of artificial‑intelligence performance while consuming far less power than conventional digital processors. The partnership, announced on February 6 2026, will license Mythic’s Analog Processing Unit (APU) technology and co‑develop a chip that is expected to power Honda’s next‑generation software‑defined vehicles (SDVs) in the late 2020s and early 2030s.
The agreement grants Honda a license to integrate Mythic’s compute‑in‑memory architecture into its vehicle platforms and establishes a co‑development roadmap that spans a five‑year term. Honda will provide the automotive‑grade design specifications, testing infrastructure, and production‑scale validation, while Mythic will supply the APU core, firmware, and silicon‑level design services. The parties have agreed on a milestone‑based payment structure that includes an upfront fee of $30 million, followed by quarterly payments tied to prototype milestones and a final production‑scale royalty of 3 % of the chip’s unit price.
Honda’s move places it in direct competition with other automakers that are pursuing in‑house or partner‑based AI chips, such as Tesla’s custom silicon, GM’s partnership with Nvidia, and Ford’s collaboration with Intel. By adopting an analog architecture that offers 100‑fold energy efficiency over digital counterparts, Honda aims to reduce the power draw of on‑board AI, thereby extending electric‑vehicle range and enabling higher‑level autonomous functions without compromising battery life.
Mythic, which recently closed a $125 million funding round led by DCVC, NEA, and SoftBank, has a track record of delivering edge‑AI solutions for consumer electronics and industrial IoT. While the company has not yet released a dedicated automotive chip, its analog technology has been validated in high‑performance, low‑power workloads that are directly applicable to vehicle perception and decision‑making systems.
The co‑developed SoC will feature a 100,000‑TOPS analog compute‑in‑memory core that operates at 1 V, delivering 100‑fold lower energy consumption than comparable digital chips. The design will meet automotive‑grade requirements for temperature extremes, vibration, and electromagnetic compatibility, and will be qualified for use in safety‑critical ADAS and autonomous driving modules.
From a strategic perspective, the partnership supports Honda’s “Safety for Everyone” mission and its goal of zero traffic‑collision fatalities by 2050. By integrating a highly efficient AI core, Honda can deploy advanced driver‑assist and autonomous features earlier while keeping vehicle power budgets in check, which is critical for its electrification roadmap that targets 100 % electric or fuel‑cell sales by 2040. "Mythic’s analog technology gives us a clear path to deliver data‑center‑grade AI performance on a vehicle‑scale platform without sacrificing range," said Atsushi Ogawa, Chief Operating Officer of Honda R&D. "This partnership is a key step toward our vision of fully autonomous, zero‑emission mobility.”
Dave Fick, co‑founder and CTO of Mythic, added, "Cars are becoming petascale supercomputers on wheels. Our analog architecture allows vehicles to run the same AI workloads that power data centers, but with a fraction of the power draw. Partnering with Honda accelerates the deployment of this technology in a high‑volume, safety‑critical environment.”
The first prototype chips are slated for integration into test vehicles by late 2028, with production‑scale manufacturing expected to begin shortly after successful validation. Honda plans to roll out the technology across its SDV portfolio, starting with the next generation of its hybrid and electric models, and eventually extending to fully autonomous platforms.
The content on EveryTicker is for informational purposes only and should not be construed as financial or investment advice. We are not financial advisors. Consult with a qualified professional before making any investment decisions. Any actions you take based on information from this site are solely at your own risk.