Microsoft launches data centers as Amazon invests $9B in Korea

Key Takeaways

• Microsoft plans to invest over $1 billion in Thailand between 2026 and 2028 to build its first data center region and enhance AI infrastructure, aiming to train millions of citizens.
• Amazon Web Services (AWS) will invest approximately $9 billion (12.6 trillion won) in South Korea by 2031 to expand its cloud infrastructure and AI services.
• New York City schools issued AI guidelines, permitting AI for tasks like brainstorming but restricting its use for student grading or counseling.
• Boulder Valley School District is developing an AI roadmap and recommends MagicSchool AI for classroom use, balancing benefits with potential dangers.
• Cleveland Clinic's Ohio locations are using AI-powered fundus cameras to screen diabetic patients for retinopathy in about 30 seconds, improving diagnosis efficiency.
• OpenAI fixed security vulnerabilities in ChatGPT and Codex that could have allowed data leaks, including user messages and GitHub tokens.
• A vulnerability in Google Cloud's Vertex AI platform, involving overly broad permissions for AI agents, was addressed after researchers found it could expose sensitive data.
• Ollama now supports Apple's MLX framework and Nvidia NVFP4, significantly boosting local AI model performance on Macs with Apple Silicon chips.
• A Silicon Valley startup is automating developer roles using AI coding tools and OpenClaw to increase efficiency and speed up software development.
• Discussions at Boston University and Marquette University highlight the need for ethical, equitable, and thoughtful integration of AI in education, addressing bias and relational impacts.

Microsoft Invests $1 Billion in Thailand for AI and Cloud Growth

Microsoft plans to invest over $1 billion in Thailand during the next two years to build cloud and artificial intelligence infrastructure. This investment includes establishing the company's first data center region in Thailand, which will serve customers across Southeast Asia. Microsoft CEO Satya Nadella announced the plan during a visit to Bangkok, emphasizing its commitment to Thailand's digital transformation. The investment aims to create jobs, develop local talent, and boost the country's digital economy.

Microsoft Boosts Thailand's AI and Cloud with $1 Billion Investment

Microsoft will invest more than $1 billion in Thailand's cloud and AI infrastructure between 2026 and 2028. This initiative, part of Microsoft's 'Advancing National Growth, Prosperity, and Global Competitiveness with AI,' aims to provide trusted access to technology and train millions of Thai citizens. The investment supports Thailand's goal to become a regional leader in the digital and AI economy. Microsoft is partnering with Thai companies like Gulf Development and CP Group to build data centers and will expand AI skills training programs.

AWS Invests $9 Billion in South Korea's AI Cloud by 2031

Amazon Web Services (AWS) will invest approximately $9 billion (12.6 trillion won) in South Korea's artificial intelligence cloud sector by 2031. This investment will expand AWS's cloud infrastructure, enhance AI services, and support local businesses in adopting AI. The plan includes building new data centers and upgrading existing facilities to meet the demand for AI computing power. AWS CEO Adam Selipsky stated the investment will empower Korean businesses to leverage AI and cloud computing for global innovation.

NYC Schools Issue AI Guidance for Classroom Use

New York City's Department of Education has released guidance on using artificial intelligence in classrooms. The guidelines use a traffic light system to show when AI tools are appropriate. Teachers can use AI for brainstorming, scheduling, and non-sensitive communications. However, AI should not be used for student counseling, grading, or handling private data. The United Federation of Teachers supports the thoughtful implementation of AI as a tool in schools.

Boulder Valley School District Creates AI Roadmap for Classes

The Boulder Valley School District is developing an AI roadmap to integrate artificial intelligence into its classes. The district encourages teachers to use the data-protected software MagicSchool AI. Superintendent Rob Anderson aims to balance the benefits of AI with its potential dangers for students and educators. The district views AI as a necessary skill for students' future and is exploring ways to safely introduce AI tools. Teachers have widely adopted MagicSchool AI for various educational purposes.

BU Forum Discusses AI's Role and Equity in Education

Boston University's Wheelock Forum explored the impact of artificial intelligence on education, urging caution and equity in its integration. Keynote speaker Aaron Rasmussen highlighted AI's rapid advancement and the need for educators to adapt. Faculty panelists discussed concerns about AI bias, particularly for sign language users, and the ethical implications of AI in surveillance and conflict. They emphasized the importance of educators guiding AI development to ensure it enhances learning equitably and thoughtfully, rather than rushing its implementation.

AI Cameras Detect Diabetic Retinopathy in Seconds at Cleveland Clinic

Cleveland Clinic's Ohio locations now use AI-powered fundus cameras to screen diabetic patients for diabetic retinopathy (DR). These cameras provide results in about 30 seconds, making annual screenings more convenient. The technology detects DR without pupil dilation, allowing screenings during regular primary care or endocrinology visits. This speeds up diagnosis and helps reserve specialist appointments for patients needing treatment. The AI software evaluates images for DR and sends results directly to the patient's electronic medical record.

OpenAI Patches ChatGPT and Codex Security Flaws

OpenAI has fixed critical security vulnerabilities in ChatGPT and Codex that could have led to data leaks. One flaw allowed attackers to secretly extract user messages and uploaded documents from ChatGPT using a DNS-based covert channel. Another vulnerability in Codex could have enabled theft of GitHub tokens. While no exploitation is confirmed, these incidents highlight risks in AI systems. Experts advise organizations to implement layered security measures when using AI tools with sensitive data.

Ollama Speeds Up Local AI Models on Macs with MLX Support

Ollama, a system for running large language models locally, now supports Apple's MLX framework for machine learning on Macs. This update, along with improved caching and Nvidia NVFP4 support, significantly boosts performance on Apple Silicon chips. Local AI models are becoming more popular as users seek alternatives to costly cloud services and desire greater privacy. While local models still lag behind frontier models, they are becoming capable enough for various tasks, offering a promising option for Mac users.

Startup Automates Developers Using AI and OpenClaw

A Silicon Valley startup is reportedly automating its developer roles using a combination of AI coding tools and OpenClaw. This technology allows for autonomous handling of tasks like code generation, debugging, and deployment. The approach aims to increase efficiency, reduce costs, and speed up software development cycles. This development signals a major shift in software creation and the future role of human developers in the tech industry.

Marquette Session Explores AI's Impact on Education Relationships

Marquette University's Center for Teaching and Learning will host a virtual session on April 7 examining artificial intelligence's role in education. The discussion will focus on the ethical, logistical, educational, and relational implications of AI. It will draw on a paper discussing how AI can either distance students from engagement or deepen connections with educators. The session aims to explore how AI can be used to foster meaningful dialogue and feedback in learning environments.

Google Cloud Vertex AI Vulnerability Exposed Sensitive Data

Cybersecurity researchers discovered a vulnerability in Google Cloud's Vertex AI platform that could allow AI agents to access sensitive data. The issue involved overly broad permissions for Per-Project, Per-Product Service Agents (P4SA), enabling AI agents to exfiltrate data and compromise cloud environments. Researchers could access sensitive data in Google Cloud Storage buckets and even download proprietary code from private Artifact Registry repositories. Google has since addressed the vulnerability, emphasizing the need for least privilege access controls for AI agents.

Extraterrestrial Intelligence Offers Advantages Over AI

The author argues that extraterrestrial intelligence (ETI) has advantages over artificial intelligence (AI) due to its training data encompassing the universe beyond Earth. Focusing solely on Earth limits human potential and survival. The essay suggests that exploring space and potentially finding ETI could guide humanity toward long-term survival, possibly through constructing interstellar spaceships. Discovering advanced ETI could inspire humanity to become an interstellar species, ensuring survival against future catastrophes.

Sources

• Microsoft to Invest Over $1 Billion in Thailand for Cloud, AI Infrastructure
• Microsoft plans $1B investment in Thailand’s cloud and AI infrastructure
• AWS boosts South Korea with 12.6 trillion won AI-cloud investment by 2031
• Parent group reacts to DOE artificial intelligence guidance
• Boulder Valley plans AI roadmap as district looks to incorporate the technology in classes
• BU Wheelock Forum Explores AI in Education
• AI-Powered Fundus Cameras Diagnose Diabetic Retinopathy in Seconds
• OpenAI Fixes Critical ChatGPT Data Leak Flaw & Codex GitHub Token Vulnerability, Raising Broader AI Security Concerns
• Running local models on Macs gets faster with Ollama's MLX support
• Meet the Startup That Used AI and OpenClaw to Automate Its Own Developers
• Education and relationships in the age of AI, April 7 
• Vertex AI Vulnerability Exposes Google Cloud Data and Private Artifacts
• Benefits of Extraterrestrial Intelligence Over AI