Author: lethuha

📌 Conclusion: Salesforce CEO Marc Benioff stated that AI-related suicides among children are the darkest aspect of this technology. He called for a reform of Section 230 of the US Communications Decency Act of 1996, which currently shields tech companies from liability for user-generated content. Without clear guardrails and accountability, friendly and personalized AI could lead to severe social consequences far outweighing technological benefits.

📌 Conclusion: The US FDA and European EMA announced a joint set of principles for “Good AI Practice” in drug development in January 2026. For the first time, the US and Europe have established a common framework for AI in pharmaceuticals, focusing on 10 pillars ranging from ethics and data to lifecycle management. This risk-based approach and GxP standards balance innovation with patient safety. This move not only accelerates drug development but also solidifies the global leadership of the EU-US in the generative AI race for biomedicine, while significantly reducing animal testing and post-market risks.

📌 Conclusion: DBS Bank recorded a record economic value from AI of 1 billion SGD in 2025. This reflects a trend across Singapore’s top three banks as they transition from generative AI to agentic AI—models capable of autonomous reasoning and execution. While expected to generate massive value, this shift raises employment concerns, prompting a massive retraining initiative for 35,000 employees.

📌 Conclusion: West Midlands Chief Constable Craig Guildford has officially apologized to British MPs after admitting to providing false information regarding the decision to ban Maccabi Tel Aviv fans. This error originated from the use of Microsoft Copilot, a generative AI tool, during the synthesis process.

📌 Conclusion: AI is not obliterating professions, but it is forcing workers to redefine their value. Data proves that soft skills—from collaboration and creativity to problem-solving—are the factors that build “AI resistance” in a career. When code and formulas become common commodities, sustainable advantage lies in the ability to connect with people, lead ideas, and transform technology into truly meaningful solutions.

📌 Conclusion: AI brings clear efficiency, but completely erasing mundane tasks can backfire. When employees are forced to maintain high intensity all day, the risk of burnout increases and creativity declines. Instead of filling every minute with performance, businesses need to actively design “white space” to allow humans to be bored and to think, thereby sparking valuable ideas.

📌 Conclusion: China’s new AI plan for manufacturing reveals a shift from a “model race” to controlling the entire AI supply chain—from chips and data to infrastructure and industrial applications. With targets of 3–5 large models, 100 datasets, and 500 scenarios by 2027, Beijing seeks both technological self-reliance and global influence. Amid U.S.–China tensions, AI is increasingly viewed as a strategic pillar on par with energy or semiconductors.

📌 Conclusion: Singapore is deploying a massive AI retraining program for all 35,000 domestic employees of DBS, OCBC, and UOB within the next 1–2 years. The focus is on agentic AI, enabling models to act autonomously and handle complex multi-step processes beyond previous AI capabilities. The government is supporting up to 90% of salaries to retrain mid-career workers.

📌 Conclusion: JLL’s report indicates that AI is not creating a bubble but is pulling the data center industry into a long-term growth cycle. With a 14% CAGR between 2026 and 2030, 100 GW of new capacity will be generated, matching the current 100 GW total. Starting in 2027, inference will surpass training, forcing data centers to be more decentralized to serve regional users. Power supply becomes the decisive factor for location, driving the “bring your own power” model, battery storage, and energy investment alongside real estate.

📌 Conclusion: 2025 shows that AI has not drastically changed total employment: the proportion of jobs with high AI exposure remains stable, and wages in this group have even increased. AI-first companies require proof that AI cannot do the job before hiring new staff; junior-level jobs in easily automated professions (such as coding and customer service) have significantly decreased post-ChatGPT, while experienced personnel roles remain stable or have increased. The concept of “workslop” has emerged, referring to AI-generated content that looks plausible but lacks depth. AI reduces thought and hiring friction, leading to the spread of superficial documents and a flood…