Semiconductors

AI200 and AI250 set a rack-scale inference push from Qualcomm

Qualcomm unveiled AI200 and AI250 data-centre accelerators aimed at high-throughput, low-TCO generative AI inference. AI200 targets rack-level deployment with high performance per pound per watt and 768 GB LPDDR per card for large models.

AI250 introduces a near-memory architecture that boosts adequate memory bandwidth by over tenfold while lowering power draw. Qualcomm pitches the design for disaggregated serving, improving hardware utilisation across large fleets.

Both arrive as full racks with direct liquid cooling, PCIe for scale-up, Ethernet for scale-out, and confidential computing. Qualcomm quotes around 160 kW per rack for thermally efficient, dense inference.

A hyperscaler-grade software stack spans apps to system software with one-click onboarding of Hugging Face models. Support covers leading frameworks, inference engines, and optimisation techniques to simplify secure, scalable deployments.

Commercial timing splits the roadmap: AI200 in 2026 and AI250 in 2027. Qualcomm commits to an annual cadence for data-centre inference, aiming to lead in performance, energy efficiency, and total cost of ownership.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

AMD powers US AI factory supercomputers for national research

The US Department of Energy and AMD are joining forces to expand America’s AI and scientific computing power through two new supercomputers at Oak Ridge National Laboratory.

Named Lux and Discovery, the systems will drive the country’s sovereign AI strategy, combining public and private investment worth around $1 billion to strengthen research, innovation, and security infrastructure.

Lux, arriving in 2026, will become the nation’s first dedicated AI factory for science.

Built with AMD’s EPYC CPUs and Instinct GPUs alongside Oracle and HPE technologies, Lux will accelerate research across materials, medicine, and advanced manufacturing, supporting the US AI Action Plan and boosting the Department of Energy’s AI capacity.

Discovery, set for deployment in 2028, will deepen collaboration between the DOE, AMD, and HPE. Powered by AMD’s next-generation ‘Venice’ CPUs and MI430X GPUs, Discovery will train and deploy AI models on secure US-built systems, protecting national data and competitiveness.

It aims to deliver faster energy, biology, and national security breakthroughs while maintaining high efficiency and open standards.

AMD’s CEO, Dr Lisa Su, said the collaboration represents the best public-private partnerships, advancing the nation’s foundation for science and innovation.

US Energy Secretary Chris Wright described the initiative as proof that America leads when government and industry work together toward shared AI and scientific goals.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Qualcomm and HUMAIN power Saudi Arabia’s AI transformation

HUMAIN and Qualcomm Technologies have launched a collaboration to deploy advanced AI infrastructure in Saudi Arabia, aiming to position the Kingdom as a global hub for AI.

Announced ahead of the Future Investment Initiative conference, the project will deliver the world’s first fully optimised edge-to-cloud AI system, expanding Saudi Arabia’s regional and global inferencing services capabilities.

In 2026, HUMAIN plans to deploy 200 megawatts of Qualcomm’s AI200 and AI250 rack solutions to power large-scale AI inference services.

The partnership combines HUMAIN’s regional infrastructure and full AI stack with Qualcomm’s semiconductor expertise, creating a model for nations seeking to develop sovereign AI ecosystems.

However, the initiative will also integrate HUMAIN’s Saudi-developed ALLaM models with Qualcomm’s AI platforms, offering enterprise and government customers tailor-made solutions for industry-specific needs.

The collaboration supports Saudi Arabia’s strategy to drive economic growth through AI and semiconductor innovation, reinforcing its ambition to lead the next wave of global intelligent computing.

Qualcomm’s CEO Cristiano Amon said the partnership would help the Kingdom build a technology ecosystem to accelerate its AI ambitions.

HUMAIN CEO Tareq Amin added that combining local insight with Qualcomm’s product leadership will establish Saudi Arabia as a key player in global AI and semiconductor development.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Gigawatt-scale AI marks Anthropic’s next compute leap

Anthropic will massively expand on Google Cloud, planning to deploy up to 1 million TPUs and bring well over a gigawatt online in 2026. The multiyear investment totals tens of billions to accelerate research and product development.

Google Cloud CEO Thomas Kurian said Anthropic’s move reflects TPUs’ price-performance and efficiency, citing ongoing innovations and the seventh-generation ‘Ironwood’ TPU. Google will add capacity and drive further efficiency across its accelerator portfolio.

Anthropic now serves over 300,000 business customers, with large accounts up nearly sevenfold year over year. Added compute will meet demand while enabling deeper testing, alignment research, and responsible deployment at a global scale.

CFO Krishna Rao said the expansion keeps Claude at the frontier for Fortune 500s and AI-native startups alike. Increased capacity ensures reliability as usage and mission-critical workloads grow rapidly.

Anthropic’s diversified strategy spans Google TPUs, Amazon Trainium, and NVIDIA GPUs. It remains committed to Amazon as its primary training partner, including Project Rainier’s vast US clusters, and will continue investing to advance model capabilities.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

South Korea moves to lead the AI era with OpenAI’s economic blueprint

Poised to become a global AI powerhouse, South Korea has the right foundations in place: advanced semiconductor production, robust digital infrastructure, and a highly skilled workforce.

OpenAI’s new Economic Blueprint for Korea sets out how the nation can turn those strengths into broad, inclusive growth through scaled and trusted AI adoption.

The blueprint builds on South Korea’s growing momentum in frontier technology.

Following OpenAI’s first Asia–Pacific country partnership, initiatives such as Stargate with Samsung and SK aim to expand advanced memory supply and explore next-generation AI data centres alongside the Ministry of Science and ICT.

A new OpenAI office in Seoul, along with collaboration with Seoul National University, further signals the country’s commitment to becoming an AI hub.

A strategy that rests on two complementary paths: building sovereign AI capabilities in infrastructure, data governance, and GPU supply, while also deepening cooperation with frontier developers like OpenAI.

The aim is to enhance operational maturity and cost efficiency across key industries, including semiconductors, shipbuilding, healthcare, and education.

By combining domestic expertise with global partnerships, South Korea could boost productivity, improve welfare services, and foster regional growth beyond Seoul. With decisive action, the nation stands ready to transform from a fast adopter into a global standard-setter for safe, scalable AI systems.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

OpenAI outlines Japan’s AI Blueprint for inclusive economic growth

A new Japan Economic Blueprint released by OpenAI sets out how AI can power innovation, competitiveness, and long-term prosperity across the country. The plan estimates that AI could add more than ¥100 trillion to Japan’s economy and raise GDP by up to 16%.

Centred on inclusive access, infrastructure, and education, the Blueprint calls for equal AI opportunities for citizens and small businesses, national investment in semiconductors and renewable energy, and expanded lifelong learning to build an adaptive workforce.

AI is already reshaping Japanese industries from manufacturing and healthcare to education and public administration. Factories reduce inspection costs, schools use ChatGPT Edu for personalised teaching, and cities from Saitama to Fukuoka employ AI to enhance local services.

OpenAI suggests that the focus of Japan on ethical and human-centred innovation could make it a model for responsible AI governance. By aligning digital and green priorities, the report envisions technology driving creativity, equality, and shared prosperity across generations.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Crypto hiring snaps back as AI cools

Tech firms led crypto’s hiring rebound, adding over 12,000 roles since late 2022, according to A16z’s State of Crypto 2025. Finance and consulting contributed 6,000, offsetting talent pulled into AI after ChatGPT’s debut. Net, crypto gained 1,000 positions as workers rotated in from tech, fintech, and education.

The recovery tracks a market turn: crypto capitalisation topping US$4T and new Bitcoin highs. A friendlier US policy stance on stablecoins and digital-asset oversight buoyed sentiment. Institutions from JPMorgan to BlackRock and Fidelity widened offerings beyond pilots.

Hiring is diversifying beyond developers toward compliance, infrastructure, and product. Firms are moving from proofs of concept to production systems with clearer revenue paths. Result: broader role mix and steadier talent pipelines.

A16z contrasts AI centralisation with crypto’s open ethos. OpenAI/Anthropic dominate AI-native revenue; big clouds hold most of the infrastructure share; NVIDIA leads GPUs. Crypto advocates pitch blockchains as a counterweight via verifiable compute and open rails.

Utility signals mature, too. Stablecoins settled around US$9T in 12 months, up 87% year over year. That’s over half of Visa’s annual volume and five times that of PayPal’s.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Europa chip by Axelera targets NVIDIA’s grip on AI accelerators

Axelera AI has introduced Europa, a new processor built to run modern AI apps on everything from small edge devices to full servers. It focuses on practical speed and low power use. The aim is to offer NVIDIA-rivaling performance without data centre-level budgets.

Inside are eight AI cores that do the heavy lifting, positioned to challenge NVIDIA’s lead in real-world inference. Helper processors handle setup and cleanup so the main system isn’t slowed down. A built-in video decoder offloads common media jobs.

Europa pairs fast on-chip memory with high-bandwidth external memory to cut common AI slowdowns. Axelera says this beats NVIDIA on speed per watt and per dollar in everyday inference. The payoff is cooler, smaller, more affordable deployments.

It ships as a tiny 35×35 mm module or as PCIe accelerator cards that scale up. That’s the same slot where NVIDIA cards often sit today. A built-in secure enclave protects sensitive data.

Research and industry partners are lining up pilots, casting Europa as a real NVIDIA rival. Early names include SURF, Cineca, Ultralytics, Advantech, SECO, Multiverse Computing, and E4. Axelera targets the first half of 2026 for chips and cards.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

USB inventor and Phison CEO warns of an AI storage crunch

Datuk Pua Khein-Seng, inventor of the single-chip USB flash drive and CEO of Phison, warns that AI machines will generate 1,000 times more data than humans. He says the real bottleneck isn’t GPUs but memory, foreshadowing a global storage crunch as AI scales.

Speaking at GITEX Global, Pua outlined Phison’s focus on NAND controllers and systems that can expand effective memory. Adaptive tiering across DRAM and flash, he argues, will ease constraints and cut costs, making AI deployments more attainable beyond elite data centres.

Flash becomes the expansion valve: DRAM stays scarce and expensive, while high-end GPUs are over-credited for AI cost overruns. By intelligently offloading and caching to NAND, cheaper accelerators can still drive useful workloads, widening access to AI capacity.

Cloud centralisation intensifies the risk. With the US and China dominating the AI cloud market, many countries lack the capital and talent to build sovereign stacks. Pua calls for ‘AI blue-collar’ skills to localise open source and tailor systems to real-world applications.

Storage leadership is consolidating in the US, Japan, Korea, and China, with Taiwan rising as a fifth pillar. Hardware strength alone won’t suffice, Pua says; Taiwan must close the AI software gap to capture more value in the data era.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Netherlands and China in talks to resolve Nexperia dispute

The Dutch Economy Minister has spoken with his Chinese counterpart to ease tensions following the Netherlands’ recent seizure of Nexperia, a major Dutch semiconductor firm.

China, where most of Nexperia’s chips are produced and sold, reacted by blocking exports, creating concern among European carmakers reliant on its components.

Vincent Karremans said he had discussed ‘further steps towards reaching a solution’ with Chinese Minister of Commerce Wang Wentao.

Both sides emphasised the importance of finding an outcome that benefits Nexperia, as well as the Chinese and European economies.

Meanwhile, Nexperia’s China division has begun asserting its independence, telling employees they may reject ‘external instructions’.

The firm remains a subsidiary of Shanghai-listed Wingtech, which has faced growing scrutiny from European regulators over national security and strategic technology supply chains.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!