Semiconductors

ByteDance prepares major AI investment for 2026

ByteDance plans a major jump in AI spending next year as global chip access remains uncertain. The firm is preparing heavier investment in processors and infrastructure to support demanding models across its apps and cloud platforms.

The company is budgeting nearly nine billion pounds for AI chips despite strict US export rules. A potential trial purchase of Nvidia H200 hardware could expand its computing capacity if wider access is approved for Chinese firms.

Rivals in the US continue to outspend ByteDance, with large tech groups pouring hundreds of billions into data centres. Chinese platforms face tighter limits and are developing models that run efficiently with fewer resources.

ByteDance’s consumer AI ecosystem keeps accelerating, led by its Doubao chatbot and growing cloud business. Private ownership gives the firm flexibility to invest aggressively while placing AI at the heart of its long-term strategy.

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

Nvidia seeks China market access as US eases AI chip restrictions

The US tech giant NVIDIA has largely remained shut out of China’s market for advanced AI chips, as US export controls have restricted sales due to national security concerns.

High-performance processors such as the H100 and H200 were barred, forcing NVIDIA to develop downgraded alternatives tailored for Chinese customers instead of flagship products.

A shift in policy emerged after President Donald Trump announced that H200 chip sales to China could proceed following a licensing review and a proposed 25% fee. The decision reopened a limited pathway for exporting advanced US AI hardware, subject to regulatory approval in both Washington and Beijing.

If authorised, the H200 shipments would represent the most powerful US-made AI chips permitted in China since restrictions were introduced. The move could help NVIDIA monetise existing H200 inventory while easing pressure on its China business as it transitions towards newer Blackwell chips.

Strategically, the decision may slow China’s push for AI chip self-sufficiency, as domestic alternatives still lag behind NVIDIA’s technology.

At the same time, the policy highlights a transactional approach to export controls, raising uncertainty over long-term US efforts to contain China’s technological rise.

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

Civil servants and AI will work together in 2050

Public administrations worldwide are facing unprecedented change as AI reshapes automation, procurement, and decision-making. Governments must stay flexible, open, and resilient, preparing for multiple futures with foresight, continuous learning, and adaptability.

During World Futures Day, experts from the SPARK-AI Alliance and representatives from governments, academia, and the private sector explored four potential scenarios for public service in 2050.

Scenarios ranged from human-centred administrations that reinforce trust, to algorithmic bureaucracies focused on oversight, agentic administrations with semi-autonomous AI actors, and data-eroded futures that require renewed governance of poor-quality data.

Key insights highlighted the growing importance of anticipatory capacity, positioning AI as a ‘co-worker’ rather than a replacement, and emphasising the need to safeguard public trust.

Civil servants will increasingly focus on ethical reasoning, interpretation of automated processes, and cross-disciplinary collaboration, supported by robust accountability and transparent data governance.

The SPARK-AI Alliance has launched a Working Group on the Future of Work in the Public Sector to help governments anticipate and prepare for change. Its focus will be on building resilient public administrations, evolving civil-service roles, and maintaining trust in AI-enabled governance.

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

Amazon considers 10 billion investment in OpenAI

Amazon is reportedly considering a $10 billion investment in OpenAI, highlighting its growing focus on the generative AI market. The investment follows OpenAI’s October restructuring, giving it more flexibility to raise funds and form new tech partnerships.

OpenAI has recently secured major infrastructure agreements, including a $38 billion cloud computing deal with Amazon Web Services (AWS). Deals with Nvidia, AMD, and Broadcom boost OpenAI’s access to computing power for its AI development.

Amazon has invested $8 billion in Anthropic and continues developing AI hardware through AWS’s Inferentia and Trainium chips. The move into OpenAI reflects Amazon’s strategy to expand its influence across the AI sector.

OpenAI’s prior $13 billion Microsoft exclusivity has ended, enabling it to pursue new partnerships. The combination of fresh funding, cloud capacity, and hardware support positions OpenAI for continued growth in the AI industry.

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

The limits of raw computing power in AI

As the global race for AI accelerates, a growing number of experts are questioning whether simply adding more computing power still delivers meaningful results. In a recent blog post, digital policy expert Jovan Kurbalija argues that AI development is approaching a critical plateau, where massive investments in hardware produce only marginal gains in performance.

Despite the dominance of advanced GPUs and ever-larger data centres, improvements in accuracy and reasoning among leading models are slowing, exposing what he describes as an emerging ‘AI Pareto paradox’.

According to Kurbalija, the imbalance is striking: around 80% of AI investment is currently spent on computing infrastructure, yet it accounts for only a fraction of real-world impact. As hardware becomes cheaper and more widely available, he suggests it is no longer the decisive factor.

Instead, the next phase of AI progress will depend on how effectively organisations integrate human knowledge, skills, and processes into AI systems.

That shift places people, not machines, at the centre of AI transformation. Kurbalija highlights the limits of traditional training approaches and points to new models of learning that focus on hands-on development and deep understanding of data.

Building a simple AI tool may now take minutes, but turning it into a reliable, high-precision system requires sustained human effort, from refining data to rethinking internal workflows.

Looking ahead to 2026, the message is clear. Success in AI will not be defined by who owns the most powerful chips, but by who invests most wisely in people.

As Kurbalija concludes, organisations that treat AI as a skill to be cultivated, rather than a product to be purchased, are far more likely to see lasting benefits from the technology.

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

Google supports UK quantum innovation push

UK researchers will soon be able to work with Google’s advanced quantum chip Willow through a partnership with the National Quantum Computing Centre. The initiative aims to help scientists tackle problems that classical computers cannot solve.

The agreement will allow academics to compete for access to the processor and collaborate with experts from both organisations. Google hopes the programme will reveal practical uses for quantum computing in science and industry.

Quantum technology remains experimental, yet progress from Google, IBM, Amazon and UK firms has accelerated rapidly. Breakthroughs could lead to impactful applications within the next decade.

Government investment has supported the UK’s growing quantum sector, which hosts several cutting-edge machines. Officials estimate the industry could add billions to the UK economy as real-world uses emerge.

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

Taiwan strengthens its role in global semiconductors

Taiwan will continue to produce the world’s most advanced semiconductors domestically to remain a vital player globally. Deputy Foreign Minister Francois Chih-chung Wu said the island’s expertise cannot be easily replicated abroad.

Taiwan has invested in fabs in the US, Japan and Germany, but warned that moving production overseas is complex. The island plans to foster international partnerships while maintaining core technology in-house to safeguard its supply chains.

China’s military pressure on Taiwan has increased concerns over regional stability and global chip supply. Wu emphasised that preventing conflict is the most effective way to secure the semiconductor industry.

Washington and Europe share strategic interests with Taiwan, including the semiconductor industry and navigation in the Taiwan Strait. Wu expressed confidence that the international community would defend these interests, maintaining Taiwan’s essential role in technology.

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

EU supports Germany’s semiconductor expansion

The European Commission has approved €623 million in German support for two first-of-a-kind semiconductor factories in Dresden and Erfurt.

A funding that will help GlobalFoundries expand its site to create new wafer capacity and will assist X-FAB in building an open foundry designed for advanced micro-electromechanical systems.

Both projects aim to increase Europe’s strategic autonomy in chip production, rather than allowing dependence on non-European suppliers to deepen.

The facility planned by GlobalFoundries will adapt technologies developed under the IPCEI Microelectronics and Communication Technologies framework for dual-use needs in aerospace, defence and critical infrastructure.

The manufacturing process will take place entirely within the EU to meet strict security and reliability demands. X-FAB’s project will offer services that European firms, including start-ups and small companies, currently source from abroad.

A new plant that is expected to begin commercial operation by 2029 and will introduce manufacturing capabilities not yet available in Europe.

In return for public support, both companies will pursue innovation programmes, strengthen cross-border cooperation, and apply priority-rated orders during supply shortages, in line with the European Chips Act.

They will also develop training schemes to expand the pool of skilled workers, rather than relying on the limited existing capacity. Each company has committed to seeking recognition for its facilities as Open EU Foundries.

The Commission concluded that the aid packages comply with the EU State aid rules because they encourage essential economic activity, show apparent incentive effects and remain proportionate to funding gaps identified during assessment.

These measures form part of Europe’s broader shift toward a more resilient semiconductor ecosystem and follow earlier decisions supporting similar investments across member states.

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

Big Tech boosts India’s AI ambitions amid concerns over talent flight and limited infrastructure

Major announcements from Microsoft ($17.5bn) and Amazon (over $35bn by 2030) have placed India at the centre of global AI investment trends, offering momentum at a time when analysts frame Indian markets as a ‘hedge’ against a potential global AI bubble.

While India has rapidly adopted AI and attracted substantial funding for data centres and chip manufacturing, including a new collaboration between Intel and Tata Electronics, the country remains a follower rather than a frontrunner in sovereign AI capabilities.

India’s government is preparing to launch its first sovereign AI model, which will support more than 22 languages. Yet its $1.25 billion investment is dwarfed by France’s €117 billion and Saudi Arabia’s $100 billion AI programmes, leaving India far behind in compute availability, R&D depth, and semiconductor infrastructure.

Despite having 2.5 times the global average concentration of AI-skilled professionals, the country faces persistent talent flight due to limited high-end domestic opportunities and a lack of competitive policy incentives.

According to EY and UNCTAD, India’ punches above its weight’ relative to its economic stage, ranking among the top nations in AI talent, startup activity, and scientific publications. Still, funding gaps remain stark: Indian AI startups raised just $1.16 billion, compared to more than $100 billion in the US and nearly $10 billion in China.

India’s emerging strength lies less in foundation-model development and more in downstream AI applications, where cost-efficient tools can drive entrepreneurship and solve local challenges such as agriculture, education, and public service delivery. Apps like MahaVISTAAR, reaching over 15 million farmers, illustrate this direction.

Yet AI also poses a threat to India’s economic backbone. Analysts warn that the country’s IT services sector, which has long been a pillar of growth, is becoming increasingly vulnerable as AI automates core business functions. Underperformance in IT stocks, reduced hiring, and stagnant wages signal early disruption.

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

Trump allows Nvidia to sell chips to approved Chinese customers

US President Donald Trump has allowed Nvidia to sell H200 AI chips to approved customers in China, marking a shift in export controls. The decision also covers firms such as AMD and follows continued lobbying by Nvidia chief executive Jensen Huang.

Nvidia had been barred from selling advanced chips to Beijing, but a partial reversal earlier required the firm to pay a share of its Chinese revenues to the US government. China later ordered firms to stop buying Nvidia products, pushing them towards domestic semiconductors.

Analysts suggest the new policy may buy time for negotiations over rare earth supplies, as China dominates processing of these minerals. Access to H200 chips may aid China’s tech sector, but experts warn they could also strengthen military AI capabilities.

Nvidia welcomed the announcement, saying the decision strikes a balance that benefits American industry. Shares rose slightly after the news, although the arrangement is expected to face scrutiny from national security advocates.

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