Yazılar

Qualcomm Moves to Arm’s Latest Tech to Boost AI Chips and Rival Apple, MediaTek

Qualcomm has adopted the newest version of Arm Holdings’ chip architecture for its next generation of flagship processors, aiming to strengthen performance in artificial intelligence (AI) and better compete with Apple and MediaTek, according to sources familiar with the matter.

The decision represents a significant turning point in the relationship between the two companies after last year’s legal clash, which had raised doubts about whether Qualcomm would continue relying on Arm’s technology. The move is also expected to boost Arm’s revenue, as the company charges more for licensing its most advanced technology.

Arm’s shares rose 5% in regular trading after Reuters reported Qualcomm’s shift to its Arm v9 instruction set — the ninth generation of the company’s computing architecture.

AI PERFORMANCE AT THE CORE

Unlike previous generations, Qualcomm’s new PC and smartphone chips will incorporate Arm’s v9 architecture, which includes several enhancements tailored for AI tasks such as chatbots, image generation, and on-device learning.
Competitors MediaTek and Apple already use the v9 standard, which defines the fundamental instructions a processor can execute and determines compatibility with apps and software.

A Qualcomm spokesperson declined to confirm specific technologies but said the company’s internal CPU design team enables flexibility:

“We chose the instructions that make sense for our customers. That’s the beauty of having our own CPU design team — we can pick and choose the instructions that add value,” the company said.

LEGAL RIFTS GIVE WAY TO PRACTICALITY

The decision marks a pragmatic end to a strained chapter in Qualcomm and Arm’s relationship. The two companies had been locked in a licensing dispute after Arm threatened to revoke a key agreement, though it later withdrew the threat.

Despite the tensions, Qualcomm’s choice to stick with Arm’s latest architecture is seen as a vote of confidence in Arm’s long-term ecosystem.
Jay Goldberg, senior analyst at Seaport Research Partners, called the decision “very positive for Arm,” adding:

“These are companies that were fighting each other. Qualcomm could have gone a very different path here.”

ARM STRENGTHENS ITS POSITION

While Arm faces emerging competition from RISC-V, an open-source alternative architecture, analysts note that RISC-V remains decades behind in maturity and lacks a comparable developer community.

Because Qualcomm licenses Arm’s architecture rather than buying complete chip designs, the exact revenue impact for Arm is uncertain. However, the shift underscores Arm’s enduring dominance in mobile and AI chip design.

As AI workloads increasingly drive hardware innovation, Qualcomm’s adoption of Arm’s newest architecture signals that the next wave of chips will focus as much on intelligence and adaptability as on speed and power efficiency.

Chip Cooling Startup Corintis Raises $24 Million, Adds Intel CEO Lip-Bu Tan to Board

Swiss startup Corintis, which develops advanced liquid cooling systems for semiconductors, announced it has raised $24 million in a Series A round and added Intel CEO Lip-Bu Tan to its board. The move comes as demand surges for efficient cooling solutions amid the AI-driven boom in high-power chips.

Corintis, based in Lausanne, was valued at about $400 million after the latest funding, a source told Reuters. The company’s technology has already been tested by Microsoft, a customer, which found it up to three times more efficient than conventional cooling systems.

Unlike traditional air-based systems—or most liquid cooling solutions that only draw heat from a chip’s surface—Corintis uses microscopic liquid channels etched inside the chip itself, allowing for more efficient heat removal while reducing power and water consumption.

The company designs its cooling systems using software automation and produces cold plates—metal blocks that transfer heat from chips into circulating liquid—in Europe. These can serve as drop-in upgrades for existing setups or be integrated directly into new chips.

“Right now we are able to produce around 100,000 cold plates per year. Next year we are ramping up to around 1 million,” co-founder and CEO Remco van Erp said. He launched the company in 2022 alongside COO Sam Harrison and another co-founder after spinning out of Lausanne’s Federal Institute of Technology.

The $24 million round was led by BlueYard Capital, with participation from Founderful, Acequia Capital, Celsius Industries, and XTX Ventures. Including earlier funding, Corintis has now raised $33.4 million.

Corintis also appointed Geoff Lyon, founder and former CEO of liquid-cooling firm CoolIT, to its board. Lip-Bu Tan joined as a director before becoming Intel’s CEO in March, and also serves as chairman of venture capital firm Walden International.

The new funds will support team growth—expanding to 70 employees by year-end from 55 today—scaling up manufacturing, and opening U.S. offices to be closer to major customers.

TSMC and Chip Design Firms Use AI to Cut Energy Use in Next-Gen Chips

The chips powering artificial intelligence consume enormous amounts of electricity, but Taiwan Semiconductor Manufacturing Co (TSMC), the world’s largest contract chipmaker, unveiled new efforts on Wednesday to make them more efficient—by using AI-powered software in the chip design process.

Speaking at a Silicon Valley conference, TSMC showcased strategies it says could boost the energy efficiency of AI chips by as much as 10 times.

Nvidia’s flagship AI servers, for instance, can draw up to 1,200 watts under heavy workloads—comparable to the electricity used by 1,000 U.S. homes if run continuously. TSMC’s approach centers on a new generation of chiplet-based designs, where multiple smaller chips made with different technologies are packaged together to function as a single processor.

To enable these designs, chipmakers are increasingly turning to AI-driven software tools. Partners like Cadence Design Systems and Synopsys debuted new products on Wednesday, built in close collaboration with TSMC. These tools have shown they can outperform human engineers in solving complex design problems—and in a fraction of the time.

“That helps to max out TSMC technology’s capability, and we find this is very useful,” said Jim Chang, deputy director of TSMC’s 3DIC Methodology Group. “This thing runs five minutes while our designer needs to work for two days.”

Still, physical constraints remain. As chips scale up, moving data on and off them via traditional electrical connections is reaching its limits. New approaches, such as optical interconnects to transfer information between chips, must be made reliable enough for deployment in massive data centers.

“Really, this is not an engineering problem,” said Kaushik Veeraraghavan, an engineer at Meta’s infrastructure group during his keynote. “It’s a fundamental physical problem.”