Ren Zhengfei, the founder of Huawei, has spent the better part of a decade running a company that the United States government has tried to systematically cut off from the world’s most advanced chipmaking technology. In July 2026, he put his personal endorsement behind what Huawei is presenting as its answer to that pressure: a chip-design approach the company calls the “Tau Scaling Law,” which Ren said in remarks published internally and later reported on July 24 is “the only path forward for Huawei to break the siege.” Unlike rivals that can choose among several technological routes to keep advancing, Ren said Huawei’s approach reflects six years of continuous iteration by tens of thousands of employees to carve out what he called a maturing, viable road forward — a description that frames the effort less as an optimization and more as a survival strategy.
A Workaround for a Technology Huawei Can’t Buy
The Tau Scaling Law is Huawei’s attempt to keep pushing chip performance forward without access to extreme ultraviolet (EUV) lithography, the advanced manufacturing technology that Western export controls have kept out of China and that underpins the most competitive chipmaking processes used by companies like TSMC and Samsung. Rather than trying to replicate EUV-dependent manufacturing, Huawei’s approach leans on chip-design innovation to extract more performance from the older, less advanced fabrication processes Chinese foundries can still access — an engineering workaround built specifically around the constraint of US sanctions rather than around what would otherwise be the industry’s default path.
Ascend Chips Are Already Turning That Into Revenue
The design philosophy isn’t purely defensive positioning — it’s underwriting a concrete, fast-growing product line. Huawei has set a 2026 target of roughly $12 billion in AI chip revenue, a more than 60% jump from the $7.5 billion the company generated from AI chips in 2025. The company’s Ascend 950PR processor entered mass production in March 2026 and has already captured the majority of the year’s chip orders, with an upgraded Ascend 950DT planned for the fourth quarter. Adoption by DeepSeek for inference workloads has been cited as one of the key drivers of that growth, positioning Huawei to capture a larger share of China’s AI compute market at a moment when Nvidia’s access to the country has been restricted by US export controls. Ren has been explicit that Huawei is not trying to out-compete the world’s leading AI labs on raw model performance, framing the company’s strategy instead around embedding AI directly into industrial settings — manufacturing, energy, logistics — saying that “superior AI only makes a company powerful, but AI applications can empower a nation.”
Record Revenue, Record R&D Spending
The chip strategy sits inside a broader financial picture that shows Huawei’s post-sanctions recovery is real, not just rhetorical. The company reported group revenue of 880.9 billion yuan (about $127.8 billion) for 2025, alongside record research and development spending of 192.3 billion yuan — a figure that underscores how much of Huawei’s revenue is being funneled directly back into the kind of design-level innovation the Tau Scaling Law represents. Even so, Ren has been careful not to overstate how fully Huawei has recovered: in comments made as the company rebounded from years of sanctions pressure, he described Huawei as “still struggling,” a characterization that reads as consistent with his broader public style — acknowledging real constraints even while presenting the company’s technical workarounds in confident terms.
A Founder Who Rarely Speaks, But When He Does, It Signals Strategy
Ren founded Huawei in 1987 and has remained notoriously media-shy compared with most major Chinese tech founders, making his direct, on-record framing of the Tau Scaling Law as existential rather than incremental a notable departure from his usual reticence. That combination — a founder known for saying little, choosing to publicly stake his own credibility on a specific technical strategy — has been read inside the industry as a signal of how central the approach is to Huawei’s long-term positioning against continued US export restrictions, rather than a routine technology announcement.
What This Means for Philippine Founders
Huawei’s Tau Scaling Law is a real, working example of a company turning a hard external constraint — no legal access to EUV lithography — into a specific, funded engineering strategy rather than treating the restriction as an unsolvable ceiling, a mindset directly transferable to any Philippine hardware, telecom, or manufacturing startup that runs into supply-chain or export-control limits of its own and needs a concrete workaround rather than a wish to have the constraint removed. Ren’s explicit choice to compete on AI applications embedded into industrial settings, rather than chasing frontier model performance the way OpenAI or Anthropic do, is also a useful positioning lesson for Philippine AI startups: with a fraction of the compute and capital of US or larger Chinese labs, competing on where AI gets deployed — manufacturing, logistics, agriculture — rather than on raw model capability is a more realistic path to building a defensible business. Huawei’s Ascend chip momentum, driven in part by Chinese AI labs like DeepSeek needing an alternative to Nvidia, is also worth watching directly: as US-China chip tensions continue, any Philippine company building on cloud AI infrastructure should track whether a genuine, cost-competitive non-Nvidia compute option becomes more broadly available outside China too.
Share this article