Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
Your premier source for technology news, insights, and analysis. Covering the latest in AI, startups, cybersecurity, and innovation.
Get the latest technology updates delivered straight to your inbox.
Reach 1.1M+ subscribers via TechBuzz Press.
Send us a tip using our anonymous form.
Reach out to us on any subject.
© 2026 The Tech Buzz. All rights reserved.
Anker's Custom Thus Chip Brings On-Device AI to Headphones
Anker unveils Thus, the first compute-in-memory AI chip for consumer audio devices
PUBLISHED: Wed, Apr 22, 2026, 10:15 AM UTC | UPDATED: Mon, May 25, 2026, 1:02 PM UTC
Anker just made a bold move into custom silicon. The consumer electronics giant unveiled Thus, what it claims is the world's first neural-net compute-in-memory AI audio chip, designed to bring local AI processing to headphones, speakers, and IoT gadgets without draining battery life. The announcement puts Anker in rare company alongside Apple and Samsung as consumer brands building their own AI-optimized processors, signaling a broader shift toward on-device intelligence in everyday accessories.
Anker just threw down the gauntlet in the race to put AI everywhere. The company known for charging cables and portable batteries unveiled Thus, a custom AI processor that rewrites how small devices handle machine learning tasks. Unlike every AI chip before it, Thus uses compute-in-memory architecture, meaning it runs calculations right where data lives instead of constantly shuffling information back and forth between separate memory and processing units.
The technical breakthrough matters because it solves the biggest problem plaguing AI-powered earbuds and smart speakers – battery life. "Every AI chip built until now stores the model on one side and does the computation on the other," Anker CEO Steven Yang told The Verge. "To think, the device has to carry all those parameters across, many times per second, every single inference." That constant data movement kills batteries and generates heat, two dealbreakers for wireless earbuds.
Thus flips that model. By embedding computation directly into memory cells using compute-in-memory architecture, the chip eliminates the energy-intensive shuttle between storage and processing. The result is a processor small enough to fit in truly wireless earbuds while running complex AI models for noise cancellation, voice recognition, and spatial audio without needing cloud connectivity.
Advertisement
The announcement positions Anker alongside tech giants like Apple and Samsung in the custom silicon game. While those companies have been designing their own chips for years – Apple's M-series processors and Samsung's Exynos line – Anker's focus on AI-specific audio processing carves out a distinct niche. The company isn't trying to build general-purpose CPUs but rather specialized processors optimized for the exact workloads its products need.
Timing matters here. The consumer electronics industry is racing to add AI features to everything from toothbrushes to coffee makers, but most implementations rely on cloud processing or power-hungry general-purpose chips. Nvidia dominates data center AI chips, while Qualcomm leads in smartphone processors, but the market for ultra-low-power AI chips in accessories remains wide open.
Anker's vertical integration play mirrors what made Apple's ecosystem so powerful. By controlling both hardware design and the software running on Thus, Anker can optimize performance in ways third-party chip buyers can't. The company hasn't disclosed which products will first ship with Thus or when they'll hit shelves, but the obvious candidates are its Soundcore headphones and Eufy smart home devices.
The broader implications extend beyond Anker's product lineup. If Thus delivers on its efficiency promises, it could pressure competitors to either license the technology or accelerate their own custom chip programs. Sony, Bose, and JBL all rely on off-the-shelf processors for their premium audio products. Anker's advantage in power efficiency and AI capabilities could force industry-wide shifts in how consumer audio devices are designed.
Advertisement
What remains unclear is how Anker will monetize this investment. Custom silicon requires massive upfront R&D costs and long development cycles. The company could keep Thus exclusive to its own brands, license it to other manufacturers, or eventually spin it into a standalone chip business. CEO Yang's comments suggest Anker views this as a competitive moat rather than a revenue stream, at least initially.
The compute-in-memory approach isn't new in research labs – IBM and other chip designers have explored it for years – but Anker appears to be first to commercialize it specifically for consumer audio AI workloads. That execution gap between lab concepts and shipping products is where hardware companies either prove their engineering chops or stumble.
Anker's Thus chip represents a calculated bet that the future of consumer AI runs locally, not in the cloud. If the company can deliver meaningful battery life improvements while adding genuine AI features to sub-$200 headphones, it'll force the entire audio industry to rethink their supply chains and chip strategies. The real test comes when products ship and reviewers can measure whether Thus lives up to its compute-in-memory promises. For now, Anker has signaled it's done playing the commodity hardware game and ready to compete on proprietary technology.
Advertisement
Advertisement
May 22
May 22
May 22
May 22
May 22
May 22
