Author: lethuphuong

📌 On October 23, 2025, Alibaba Cloud announced the development of the Aegaeon system, which allows for optimal GPU utilization in concurrent AI inference tasks, helping to reduce the number of required GPUs by up to 82% while maintaining high performance. Alibaba Cloud demonstrates China’s rapid progress in optimizing GPU infrastructure for generative AI, reaching a performance level of 7 models/GPU. This technology not only reduces the cost of inferencing billions of AI requests but could also reshape the global AI cloud market, where GPU optimization capability is becoming the most crucial competitive weapon.

📌 China’s Fourth Plenum marks a strategic turning point: prioritizing AI, space technology, and population economics, while strengthening anti-corruption efforts and macro-stability. With 4,500 AI companies, the direction for the next five years shows China’s ambition to become an AI and space technology powerhouse, but still maintaining US-China dialogue to avoid global economic exposure risks.

📌 The Consensus platform uses GPT-5 and Responses API to shorten research from weeks to minutes. Using GPT-5 and Responses API, Consensus is redefining the scientific research process, turning AI into a research assistant that can read, understand, and verify evidence like a human. With processing speeds hundreds of times faster but still ensuring academic reliability, this platform opens up an era of “instant research,” where science advances faster thanks to AI’s ability to synthesize knowledge at super speed.

📌 Don’t be fooled: Contrary to the image of “laissez-faire” technology, the US government is in fact tightly controlling the core foundation of the AI ecosystem. US AI policy is one of deliberate intervention: loose at the application layer to foster innovation, but maintaining absolute control at the infrastructure layer – chips, data, and foundational models. It is a new soft power strategy, where technology becomes a political tool in the global AI hegemony competition.

📌 China’s state-owned defense conglomerate Norinco launched the P60 autonomous combat vehicle, reaching speeds of 50 km/h, operated by the DeepSeek AI model. The People’s Liberation Army (PLA) continues to seek access to Nvidia chips (A100, H100) despite the US export ban since September 2022. Many 2025 patents still report the use of these chips, but the PLA is also increasing its shift to Huawei Ascend chips to achieve “algorithmic sovereignty,” alongside researching fighting dog robots that operate in swarms, perform bomb disposal, and reconnaissance. DeepSeek can simulate 10,000 battlefield scenarios in 48 seconds, 3,600 times faster than a human…

📌 OpenAI’s report exposes an alarming scale: over 1 million people per week express suicidal intent when using ChatGPT, along with 560,000 potential cases of mental disorders. Although GPT-5 achieved a 91% safety standard, public pressure and legal investigations are forcing OpenAI to prove that generative AI can support, rather than exacerbate, the global mental health crisis.

📌 Singapore is positioning itself as a “pragmatic bridge” in the era of fragmented AI regulation – neither extreme like Europe nor centrally controlled like China. Guided by the philosophy “AI for public good – for Singapore and the world,” the nation prioritizes technical standards, social responsibility, and international competitiveness, becoming a reference model for smaller economies seeking sustainable and responsible AI development.

📌 The article “The Gen AI Playbook for Organizations” in Harvard Business Review (November–December 2025 issue) asserts: businesses cannot wait for “perfect AI”—they need an immediate generative AI adoption strategy, focusing on strategic differentiation rather than speed of deployment. Organizations should review their organizational structure and redesign processes to suit an “AI-native organization.” Personnel will shift from “operating tools” to “collaborating with AI.” Long-term advantage will come from unique AI usage, proprietary data, and a culture of human-machine collaboration.

📌 A new trend is spreading among AI startups: self-collecting proprietary data, instead of scraping the Internet or hiring cheap labor for data labeling as before. It’s the quality of the data, not the quantity, that determines performance. This is a competitive advantage that competitors find hard to replicate. This trend marks a revolution in data quality — where the most effective models are built from real data, real people, and real actions.

📌 The AI boom is reshaping the US financial sector: JPMorgan and Goldman are both choosing to reduce hiring despite quarterly profits reaching $14.4 billion and $4.1 billion respectively. The focus is on applying AI to boost efficiency, reduce personnel costs, and fundamentally re-engineer the operating model, marking a turning point in AI transformation within the banking industry.