Arteris Crossed 1,000 Chip Design Starts: AI Has Taken Half of Its Recent Projects

Arteris, the Nasdaq-listed network-on-chip IP vendor trading under the ticker AIP, said it has surpassed 1,000 customer semiconductor design starts using its interconnect intellectual property, and reported that AI-related projects accounted for 50% of its design starts in 2021-2025, up from 4% in 2010-2014. The company released "The State of Semiconductor Innovation: Insights from 1,000 Design Starts" alongside the milestone, which counts customer projects entering development rather than finished chips; data centers, cloud and other high-performance computing applications rose to 43% of design starts in 2021-2025 from 11% in 2010-2014, and Arteris reported a ninefold increase in measured design complexity over the past two years. CEO K. Charles Janac said "the biggest number of design starts is going to come from Physical AI," while cautioning that "nobody really knows where AI can be applied economically.

Arteris Crossed 1,000 Chip Design Starts: AI Has Taken Half of Its Recent Projects The company, which trades on Nasdaq under the ticker AIP, announced the achievement alongside a report examining its historical customer design activity. Arteris said it has surpassed 1,000 customer semiconductor design starts using its interconnect intellectual property, marking a milestone that illustrates how artificial intelligence https://www.ibtimes.com/topic/ai is changing the architecture of chips powering data centers, vehicles and connected devices. For CEO K. Charles Janac, the opportunity increasingly lies in moving the enormous volumes of information those systems require. "A chip basically does three things. It processes data, it stores data, and it moves data," Janac told IBT in an interview. "Our goal is to be the leader in the data movement part of the chip." The Nasdaq-listed company, which trades under the ticker AIP, develops network-on-chip NoC technology that connects components within chips and between chiplets. Its milestone counts customer projects entering development, rather than finished chips or devices shipped. Arteris released "The State of Semiconductor Innovation: Insights from 1,000 Design Starts" alongside the announcement. AI-related projects accounted for 50% of its design starts in 2021-2025, up from 4% in 2010-2014. "These AI chips need to move incredible amounts of data," Janac said, describing the traffic between memory and processing elements. Arteris has responded by developing wider data pathways and supporting more transactions and higher operating frequencies. Gina Jacobs, Arteris' vice president of brand marketing, said the underlying challenge becomes clearer when considering how smartphones work. "All of those components on a chip, whether it's the processor, the GPU, memory, they all need to talk to each other," Jacobs told IBT. "They all need to communicate and move data." Those connections influence battery life, performance and how quickly applications load, she explained. Similar demands subsequently expanded into automotive systems and AI data centers. Data centers, cloud computing and other high-performance computing applications represented 43% of Arteris design starts in 2021-2025, compared with 11% in 2010-2014, according to the report. The company also reported a ninefold increase in measured design complexity over the past two years. Asked how Arteris assessed that change, Jacobs pointed to the number of networks-on-chip in customer designs and the connections needed to move data among components. She linked that growth to increasingly demanding applications. "The more the system does, the more complex the software and the more complex is the silicon," Janac detailed. Some designs have become too complex to accommodate their functions on one chip, he added, prompting developers to distribute them across multiple chiplets connected inside a package. Automotive systems offer another indication of where demand is heading. Approximately two-thirds of automotive design starts carried a functional safety classification, and nearly half combined that classification with AI capabilities, according to Arteris. Janac said driver-assistance chips helped bring the company into AI, as vehicles needed to distinguish people, animals and obstacles from camera data. He expects related requirements to spread through physical AI, including autonomous machines interacting with their surroundings. "I think ultimately the biggest number of design starts is going to come from Physical AI," Janac said. Arteris is also seeing a "surprising number of space-based design starts," he added, citing orbital constellations and exploration vehicles. Still, Janac cautioned that technological promise does not guarantee every investment makes economic sense. "Short term, some of the investments are not particularly well placed," he said. "Nobody really knows where AI can be applied economically." At Arteris, he described efforts to use AI in programming, automate customer support, and develop future product capabilities. "You need to have your eyes open when you apply AI," Janac said, "in order to make sure that you're applying it economically." © Copyright IBTimes 2026. All rights reserved.