Tech Giants Bet Big on Neuromorphic Chips and AI Accelerators in 2025

August 25, 2025

By Silo AI Team

Tech Giants Bet Big on Neuromorphic Chips and AI Accelerators in 2025

Context and Industry Drivers

The increasing investment in neuromorphic chips and AI accelerators by tech giants in 2025 is driven by the need to address AI's growing energy demands and unlock new capabilities in edge computing, robotics, and data centers. Neuromorphic computing, which mimics the human brain's neural architecture, enables ultra-low-power, event-driven AI processing that is well-suited for edge devices and real-time applications.

The neuromorphic chip market is projected to reach $8.3 billion by 2030, fueled by the need for sustainable AI solutions and the proliferation of AI-powered IoT devices. Training large AI models like GPT-4 consumes enormous energy (up to 300,000 kWh per model), making energy efficiency a top priority for AI infrastructure.

Tech Giants and Their Bets

Tech giants are making significant investments in neuromorphic chips and AI accelerators, with varying strategies and focuses:

IBM

IBM debuted NorthPole (2023), a next-gen chip fusing memory and compute for energy-efficient AI inference. In 2025, IBM is focused on hybrid-cloud, licensing, and defense partnerships, with NorthPole currently a lab device and commercial path and scaling still under consideration.

Intel

Intel's Loihi 2 (2021, upgraded 2024) is a digital, spiking neural network chip available to research partners. Intel is positioned to scale rapidly if demand spikes, leveraging its foundry capacity.

BrainChip

BrainChip's Akida is a pure-play neuromorphic chip for ultra-low-power edge AI, used in sensors and defense. While revenue is modest, the company holds IP/licensing deals with major tech and defense firms.

Nvidia

Nvidia continues to dominate AI chip supply, with a valuation over $1 trillion in 2025. While Nvidia's main focus is high-performance GPUs for AI, it is also investing in new AI accelerators to maintain leadership.

Key Trends in 2025

Several key trends are shaping the neuromorphic chip and AI accelerator landscape in 2025:

Silo Tech Article Banner - Tech Giants Bet Big on Neuromorphic Chips and AI Accelerators in 2025

Energy Efficiency

Energy efficiency is the main motivation behind the investment in neuromorphic chips, as AI workloads scale and the cost and sustainability of powering large models become central concerns.

Edge AI

Neuromorphic chips' low power consumption is essential for IoT and edge devices, with Gartner predicting 70% of IoT endpoints to run AI by 2027.

Research and Government Funding

Major public and private investments, such as China's $10B "Made in China 2025" push for domestic AI chip R&D, are accelerating innovation and competition.

Commercialization Challenges

While tech giants have advanced lab prototypes, mass-market adoption is limited by manufacturing, ecosystem readiness, and the need for software and developer tools.

Outlook

In 2025, tech giants view neuromorphic chips as a strategic bet—not yet a replacement for GPUs in mainstream AI, but a potential inflection point for energy-constrained or edge AI applications. Most neuromorphic projects remain in research or early deployment phases, with large-scale commercialization expected later in the decade.

The race for AI accelerators includes both neuromorphic and traditional architectures, reflecting a broader industry effort to diversify and future-proof AI hardware stacks. As the industry continues to evolve, it will be interesting to see how these investments shape the future of AI computing.

For more insights on the latest developments in AI and tech, check out our articles on Global AI Policy Shakeup: New Regulations Target Data Privacy & Bias and NASA–IBM’s Surya AI Model Sets New Bar for Space Weather Forecasting.