AI Chip Costs Shift to Memory; Microsoft Cuts Third-Party AI Tools

High Bandwidth Memory (HBM) is projected to constitute 63% of AI chip component costs by Q4 2025, up from 52% in Q1 2024, driven by increased production for major AI chip manufacturers. This shift suggests that memory costs, rather than compute logic, will increasingly pressure AI infrastructure expenses. Meanwhile, Microsoft is reportedly reducing its reliance on third-party AI coding tools like Claude Code, migrating engineers to its own GitHub Copilot CLI to manage internal AI expenses, highlighting a growing concern over the total cost of AI agent implementation. AI

IMPACT Rising HBM costs will increase AI infrastructure expenses, while companies like Microsoft are re-evaluating AI tool adoption to control operational costs.

RANK_REASON The cluster discusses significant shifts in AI chip component costs and major companies' strategies for managing AI tool expenses. [lever_c_demoted from significant: ic=1 ai=0.7]

Read on dev.to — LLM tag →

• Amazon
• AMD
• ChatX-One
• Claude Code
• DeepSeek
• GitHub Copilot CLI
• Google
• Hancom
• KanBots
• LG AI Research
• Microsoft
• Nvidia
• Reasonix
• Uber

• infra
• product
• policy

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →