Arm’s AI Platform for Startups

Arm has announced the availability of its most powerful AI platform for edge devices, Armv9, to startups through its Flexible Access program. This program offers an ideal opportunity for companies to experiment with the technology before purchasing licenses, enabling innovation and development without upfront costs.

Flexible Access Program: Innovation Without Risk

The “Flexible Access” model serves as a “try before you buy” system for chip designers. This model provides companies with access to a wide range of Arm technologies, tools, and resources at low or no cost for eligible startups. Companies can freely experiment and modify their designs, paying licensing fees only for the technology used in the final designs.

This approach has proven to be a “catalyst for innovation,” contributing to the creation of approximately 400 successful chip designs over the past five years. Among the companies that have benefited from this program are Raspberry Pi, Hailo, and SiMa.ai.

Armv9 Platform: Processing Power and Autonomy

The Armv9 platform features the highly efficient Arm Cortex-A320 processor, working alongside the Arm Ethos-U85 neural processing unit. This unit handles heavy AI tasks and is capable of running AI models with over a billion parameters directly on the device, without needing cloud connectivity.

This technology will support future smart applications at the edge, such as smart cameras that not only record but understand what they see, smart home devices that learn from user habits, and robots that can interact using vision, sound, and gestures.

Privacy and Security in Local Processing

One of the key benefits of the Armv9 platform is privacy and security. By processing data locally, devices can interact and respond like humans while keeping inference and data processing secure on the device. There is no longer a need to send personal data to an external server for analysis.

The Armv9 platform includes security features such as Pointer Authentication Code (PAC) and Memory Tagging Extension (MTE) to ensure data security on the device.

Future Prospects of AI in IoT

Research from VDC indicates that AI will become the “dominant technology” in IoT projects by 2028. Arm technology is already playing a pivotal role in this transformation, and this announcement further strengthens its leadership position in the field.

For developers eager to start working with AI at the edge, the Arm Cortex-A320 processor will be available through the program in November 2025, with the AI Ethos-U85 processor available in early 2026.

Conclusion

The Armv9 platform is a significant step towards enhancing AI technology at the edge, offering startups the opportunity to innovate without substantial financial risk. With the anticipated shift in IoT projects towards AI, Arm remains at the forefront of this transformation, ensuring a future led by smart and secure technology.