More than half of China's AI token usage is reportedly concentrated in short dramas and video generation, but ByteDance's Doubao has disputed the figure, saying video-generation models account for only about 10% of overall computing resources. The conflicting estimates highlight the difficulty of assessing AI compute consumption and its commercial value.

Meta Platforms has launched approximately US$35 billion of concurrent underwritten public offerings and plans an approximately US$60 billion at-the-market (ATM) program, as it funds capex it expects to reach US$135 billion to US$145 billion in 2026. The company said the equity raises complement operating cash flow, investment-grade debt, and strategic partnerships, and will further diversify its funding sources.

Nvidia opened GTC 2026 in Berlin on October 21 with a keynote highlighting AI infrastructure, security architecture, and Physical AI, while a Greenpeace protest briefly interrupted proceedings. A robot demonstration also exposed limits in motion control, instruction comprehension, and execution stability as Europe weighs innovation, energy demand, regulation, and competitiveness.

AWS has launched the open-source TorchNeuron tool, allowing developers to port PyTorch artificial intelligence (AI) applications to its custom Trainium chips without rewriting code. Google has followed with TorchTPU, while AMD and Intel are integrating AI software optimizations as chipmakers challenge Nvidia's market advantage built on the CUDA ecosystem.

India's tech push opens markets unevenly: foreign suppliers get first crack at Tata's fab, Lam courts local vendors, Jio's IPO tests its 5G pitch, and CleanMax powers AI data centers.

Chunghwa Telecom (CHT) has established a new subsidiary, CHT Intelligent Vision Solutions (IVS), spinning off an internal team focused on smart security and intelligent video analytics. The new company will target enterprise security, public safety, and smart-city applications, while taking Taiwanese hardware partners overseas.

Concerns over a potential AI bubble are intensifying. From large language models (LLMs) and AI data centers to humanoid robots, investors are increasingly questioning when massive capital investments will translate into corresponding revenue and profits.

Nvidia has won a CPU supply order from AI startup Modal, challenging Intel and AMD in the server CPU market as it expands beyond GPUs into general-purpose computing. Although Modal's annual revenue is below US$1 billion, and neither the order's value nor its deployment scale has been disclosed, the deal underscores a broader shift in AI infrastructure competition toward full-stack system integration.

Tech Mahindra has announced a collaboration with AI coding platform SpaceXSI to help global enterprises modernize legacy applications, accelerate software delivery, and improve developer productivity, according to a company announcement issued from Pune, India, on October 22, 2026.

AI-driven growth will push the semiconductor industry past the US$1.6 trillion mark in 2026, while accelerating chip development cycles to one new product every nine months, Etron Chairman Nicky Lu said. He warned that the next 3 to 5 years will be critical for Taiwan, which risks facing intense competition unless it builds stronger software, system, and architecture advantages.

Elan Microelectronics is pursuing drone opportunities across multiple applications, showcasing a broad range of solutions at the 2026 TAITRONICS expo. Solutions developed with Taiwanese drone maker Thunder Tiger stood out at the company's exhibition area, with several offerings expected to begin contributing to revenue in 2026 and 2027.

IBM's third quarter went some way towards answering whether the deal slippage that hit its second quarter reflected weaker demand or delayed timing. Revenue rose 5% year on year, or 6% at constant currency, to US$17.1 billion for the quarter ended September 30, 2026, an acceleration after a second quarter that led the company to cut its full-year outlook in July.

Meta Platforms' third quarter showed that its core advertising business is still growing fast enough to carry one of the largest infrastructure programs in the technology industry. Still, it also showed legal liabilities becoming a material cost line. Revenue rose 26% year on year to US$64.6 billion for the quarter ended September 30, 2026, while the operating margin fell to 20% from 40% a year earlier.

Grab Holdings' proposed acquisition of Delivery Hero's Foodpanda operations in Taiwan has been rejected by Taiwan's Fair Trade Commission (FTC). The commission made the announcement on October 21, blocking the deal on the grounds that it would restrict market competition. Grab issued a statement saying while it was disappointed with the decision, it will continue developing its core businesses in its existing Southeast Asian markets.

AMD and Intel plan to consolidate their PyTorch optimizations for x86 processors in oneDNN, aiming to give AI developers a common path to performance improvements across both companies' CPUs and reduce the work of maintaining separate software paths.

The US imposed sanctions on five Chinese entities tied to ex-Google engineer Linwei Ding for theft of trade secrets. The action follows his conviction for stealing confidential AI technology from Google, apparently to benefit Chinese firms.

German Chancellor Friedrich Merz and Nvidia CEO Jensen Huang urged German businesses to embrace AI at an industry summit in Berlin on October 19, aiming to revive the economy and modernize industry. Huang specifically called on Germany's automotive and industrial machinery sectors to adopt physical AI by leveraging the country's extensive manufacturing data and expertise.

Intel has asked the Federal Communications Commission (FCC) for a waiver to the US ban on advanced foreign robots, citing worries that its machines will not receive equipment updates, which could hinder its ability to produce semiconductors in the US. The request comes after the US banned imports of foreign-made robots in July.

The US Department of Homeland Security (DHS) has ordered federal civilian agencies to remove DeepSeek products and third-party services built on the Chinese AI model from their information technology (IT) systems, escalating Washington's scrutiny of Chinese AI.

Toyota plans to transition its entire lineup of approximately 50 vehicle models to software-defined vehicles (SDVs) by 2030, equipping them with advanced software capabilities that allow owners to upgrade vehicle functions through over-the-air (OTA) updates after purchase.

Singapore is moving its AI strategy beyond pilot-heavy experimentation toward a more deployment-focused model centered on enterprise adoption, sector transformation and resilient access to compute.

Nvidia is expanding its presence in China's gaming market by bringing its next-generation Deep Learning Super Sampling (DLSS) 5 neural rendering technology to titles from major Chinese developers, including NetEase, Tencent, Perfect World's Hotta Studio, and Beijing Lingyoufang Network Technology.

Japan is stepping up its push into physical AI, with major technology companies and research institutions showcasing advances in autonomous robot learning, motion sensing, and agricultural automation at CEATEC 2026.

As interest rates fluctuate, bond yields rise, and Chinese AI models intensify competition, market attention is shifting from technological development and infrastructure expansion to investment returns and long-term debt-servicing capacity. The AI industry continues to attract capital but must also prove that its business models can generate profits.

As Manus, one of China's more distinctive AI startups, became entangled in the high-profile unwinding of its acquisition by Meta, the startup's failed attempt to capture opportunities beyond China left the industry speculating about its next steps after an abrupt regulatory reversal.