This article first appeared in The Edge Malaysia Weekly on December 29, 2025 - January 4, 2026

THE global semiconductor industry in 2025 was defined by the artificial intelligence (AI) boom, volatile geopolitical manoeuvring and the intensifying battle for technological dominance between the US and China.

From regulatory shifts and trade tensions to blockbuster partnerships and surging valuations, the year revealed both the immense economic potential of semiconductors and the structural vulnerabilities of an industry at the centre of global power struggles.

The year began with a contentious policy debate in Washington.

On Jan 13, just a week before Joe Biden exited the White House, his administration introduced the now infamous AI Diffusion Rule, which was intended to tighten export controls on advanced AI chips.

Biden’s eleventh-hour move drew sharp criticism from the US semiconductor industry, including the US-based Semiconductor Industry Association (SIA) and chip giant Nvidia Corp.

Industry voices warned that the restrictions risked undermining the US’ technological leadership while curtailing revenue from international sales.

Even the charismatic Nvidia CEO Jensen Huang, who is known for his measured public statements, publicly decried the rule, calling Biden’s policy a “failure” and arguing that it threatened the company’s competitiveness in China and other markets.

The political landscape shifted dramatically with Donald Trump’s return to the White House. One of Trump’s first moves was to revoke Biden’s 2023 executive order addressing AI risks to consumers, workers and national security.

Trump’s administration then began recalibrating the US’ chip export policy by scrapping Biden’s AI Diffusion Rule in the middle of May. Huang praised the new direction and highlighted the importance of maintaining the US’ competitive edge.

In Huang’s view, if the US wants to stay at the forefront of the AI race and it wants the rest of the world to build on American technology, AI diffusion needs to be maximised.

Despite ongoing hiccups in the sale of its chips to China, Nvidia’s strong market performance reflected both the AI boom and investor confidence in its strategic positioning. On Oct 29, it made history by becoming the first company to surpass US$5 trillion (RM20.25 trillion) in market capitalisation, driven by sustained demand for its H100 and Blackwell chips.

Since the release of generative AI chatbot ChatGPT in November 2022, Nvidia’s shares have risen twelve-fold, mirroring the rapid proliferation of AI applications. The Silicon Valley firm’s dominance, however, has also placed it squarely in the cross hairs of US-China tensions.

Trump continued to restrict the sale of Nvidia’s most advanced chips to China, explicitly barring China from accessing the technology. It was also reported that the US government had been embedding trackers in AI chip shipments to prevent diversion and potential military use by Chinese entities.

Huang repeatedly cautioned against isolating China, which he notes is home to roughly half of the world’s AI researchers. Against this backdrop, Nvidia is reportedly preparing to begin shipping its H200 chips — part of the previous-generation Hopper line — to China as early as mid-February.

The move would mark the first deliveries of H200 chips to China since Trump said Washington would allow such sales, subject to a 25% fee. However, uncertainty remains, as Beijing has yet to formally approve any H200 purchases and the timeline could still shift depending on regulatory decisions.

In parallel with US policy shifts, China continued its rapid push into AI and semiconductors. DeepSeek, powered by Huawei’s 910C chipsets, emerged as a formidable competitor to ChatGPT.

Ant Group, backed by Alibaba co-founder Jack Ma, is also using Chinese-made semiconductors to develop techniques for training AI models, while Shenzhen-headquartered tech giant Tencent Holdings Ltd released a new AI model claiming to have faster responses than DeepSeek-R1.

Huang applauded the Chinese firms, especially Huawei and DeepSeek, emphasising the technological sophistication of Huawei’s chips and the competitive pressure from Chinese AI ventures.

Notably, Chinese President Xi Jinping had in February publicly voiced support for Ma and other private-sector technology leaders, reinforcing China’s intent to challenge US dominance in AI and semiconductors. Corporate observers believed this could be the end of the tech crackdown in China.

The emergence of Chinese semiconductor equipment firm Shenzhen SiCarrier Technologies Co Ltd as a new challenger to chip-tool titans such as Dutch firm ASML Holding NV and other established foreign names further illustrates China’s ambition to develop end-to-end semiconductor capabilities independent of Western technology.

Still, Trump repeatedly rattled the global semiconductor industry with threats of sweeping sectoral tariffs on semiconductors imported from countries without US-based manufacturing, first floating a 100% levy before escalating the rhetoric to a staggering 200% to 300%. As of now, no decision has been made.

Amid the turbulence, Trump on Aug 7 publicly criticised Intel Corp’s new CEO Tan Lip-Bu, citing perceived conflicts of interest linked to China. Quite bizarrely, a few days later, he took a different tone, saying he and his cabinet members had met Tan and describing his rise and success as “an amazing story”.

Malaysian-born Tan, who has been an Intel director since 2022, replaced Pat Gelsinger as CEO in March amid layoffs and market challenges. Although Tan has a monumental task in bringing Intel back to its glory days, he made a good start by persuading Nvidia and the US government to place confidence in his firm.

In September, Nvidia announced a US$5 billion investment in its struggling rival Intel, reportedly for a 4% stake. A month earlier, the US federal government had made a historic move by taking a 10% equity stake in Intel following an US$8.9 billion capital injection.

Direct government equity ownership in US private companies remains highly unusual, but Trump has signalled a willingness to make similar interventions where he deems national and industrial interests to be at stake.

On Oct 6, Advanced Micro Devices Inc (AMD) said it had signed a multi-year agreement to supply AI chips to OpenAI — a deal that could generate tens of billions of dollars in annual revenue and give the ChatGPT creator the option to acquire up to about 10% of the chipmaker. The announcement sent AMD shares soaring more than 34% that day.

A month earlier, rival Nvidia had unveiled its own agreement with OpenAI to supply at least 10 gigawatts of AI systems. Unlike OpenAI’s deal with AMD, which includes an option to take an equity stake in the chipmaker, Nvidia’s arrangement sees the chip giant committing to investing up to US$100 billion in ChatGPT’s parent, underscoring the vastly different structures.

In Asia, Nvidia cemented its presence in South Korea, supplying over 260,000 Blackwell AI chips to Samsung Electronics, Hyundai Motor Group and SK Group.

Samsung Electronics is building an AI factory with over 50,000 graphics processing units (GPUs) to accelerate its AI, semiconductor and digital transformation road map. SK Group is building an AI factory featuring over 50,000 Nvidia GPUs and Asia’s first industrial AI cloud for physical AI and robotics workloads.

Hyundai Motor Group, meanwhile, is collaborating with Nvidia and the South Korean government to build an Nvidia AI factory with 50,000 Blackwell GPUs to enable integrated AI model training, validation and deployment for manufacturing and autonomous driving.

These agreements culminated in the now-famous “chimaek summit” on Oct 30 in Seoul, where three billionaires — Nvidia’s Huang, Samsung chairman Lee Jae-yong and Hyundai boss Chung Eui-sun — dined on Korean fried chicken and beer, symbolising friendship amid complex trade negotiations.

Over the dinner, Huang presented his hosts with a bottle of Japan’s Suntory Hakushu 25-year-old single malt whisky and Nvidia’s DGX AI system, while also distributing small gifts to nearby citizens. The gathering, which drew crowds outside the restaurant, concluded with the trio sharing leftover chicken with onlookers and a playful exchange over the dinner bill, showing the human side of high-stakes global business negotiations.

As the billionaires shared that light-hearted dinner in Seoul, a more deliberate and strategic tone played out elsewhere in the global AI arena.

Tareq Amin, head of Saudi Arabia’s AI start-up Humain, said in October that he is confident the US will approve the sale of advanced chips to the kingdom after his company provided detailed assurances to US officials. Among these guarantees was a commitment that Humain would not purchase equipment from China’s Huawei.

The US has long imposed restrictions on China’s tech sector, including blacklisting firms such as Huawei and Semiconductor Manufacturing International Corp (SMIC). Humain, owned by the kingdom’s Public Investment Fund, was launched in May as the cornerstone of Saudi Arabia’s strategy to become a major force in the AI space.

Trump had from May 13 to 16 made his first major international trip of his second presidential tenure, visiting Saudi Arabia, Qatar and the United Arab Emirates. Accompanied by a Republican delegation and several business leaders, including senior adviser Elon Musk, Trump’s visit coincided with the Saudi-US Investment Forum. It was announced at the forum that Nvidia and AMD would supply advanced semiconductors to the Gulf Region.

Nvidia’s relationship with US policymakers remains central to its global strategy, as it continues to face scrutiny over its chip sales to China. In November, four individuals were charged in Florida for attempting to smuggle advanced Nvidia GPUs to China, highlighting ongoing enforcement of export controls.

Amid Nvidia’s dominance, other US chipmakers pursued strategic initiatives to remain relevant. Intel, under Tan, focused on expanding domestic capacity and navigating international challenges, including tariffs and regulatory scrutiny. AMD leveraged its OpenAI partnership to strengthen its foothold in AI accelerators and compete more aggressively.

Meanwhile, Micron Technology Inc is benefiting from a global supply crunch in memory chips amid surging demand from AI data centres. Shortages across industries — from smartphones to data centres — have driven prices higher, prompting Micron to forecast second-quarter adjusted profits that are nearly double Wall Street’s expectations. The company is one of only three major suppliers of high-bandwidth memory (HBM) chips critical for AI, alongside South Korea’s Samsung and SK Hynix.

Chinese players such as DeepSeek, Huawei and SiCarrier accelerated efforts to challenge Western dominance. Tencent and Ant’s AI advancements, combined with government backing, signalled China’s intention to secure strategic autonomy in AI and semiconductor production.

All in all, 2025 solidified the semiconductor industry’s role as a critical driver of the AI revolution and global economic strategy. Nvidia’s unprecedented valuation, AMD’s partnership with OpenAI and the rise of Chinese competitors highlighted both opportunity and risk in an increasingly strategic sector. The US-China tech rivalry, tariff negotiations and policy shifts under Trump created a dynamic landscape where geopolitical considerations were inseparable from commercial decisions.

Executives such as Huang, Lee, Chung and Tan navigated this environment with a blend of technological foresight, diplomacy and calculated risk-taking. Saudi Arabia’s Humain and China’s DeepSeek reflected the global appetite for AI capacity, while the orchestration of deals, investments and partnerships shows the importance of cross-border collaboration in an era of constrained supply chains and regulatory scrutiny.

As 2025 closes, the semiconductor industry is at a pivotal moment. Companies face the challenge of scaling AI infrastructure while managing geopolitical risk, compliance and competitive pressure. Yet the events of the year — from Nvidia’s historic market capitalisation and the emergence of new Chinese players, to strategic US policy shifts and high-profile global partnerships — suggest that 2026 will continue to be defined by rapid technological progress, intense competition and the intersection of commerce and geopolitics at the heart of the AI revolution. 

Save by subscribing to us for your print and/or digital copy.

P/S: The Edge is also available on Apple's App Store and Android's Google Play.