Latest Pulse
- Sep 23, 2026China Probes DeepSeek, Moonshot Over Claude Data RoutingChinese regulators opened a probe into DeepSeek and Moonshot AI after Anthropic accused seven Chinese companies of secretly routing customer requests through Claude to harvest its outputs, sending Alibaba, Zhipu and MiniMax shares lower.→
- Aug 18, 2026Xiaomi Profit Falls 43% as Memory Chip Costs BiteXiaomi's adjusted net income fell 43% to 6.22 billion yuan in its third consecutive quarterly profit decline, as a persistent global memory-chip shortage -- driven substantially by AI data center demand for the same DRAM and NAND -- raised smartphone component costs faster than Xiaomi could pass them on.→
- Aug 4, 2026Tesla's China Business Is Its Cash Cow and Its RiskArs Technica examines how much of Tesla's profitability still runs through China, and how quickly domestic Chinese EV makers are taking the share that funds it.→
- Jun 24, 2026Xiaomi's HarnessX Rewrites Its Own AI Scaffolding Mid-Task -- and Smaller Models Gain the MostXiaomi unveiled HarnessX, a system that lets an AI agent rewrite its own scaffolding -- the prompts, tools and control logic around the model -- in the middle of a task. The standout finding: smaller, cheaper models benefit the most, suggesting clever orchestration can substitute for raw model size.→
Coverage Timeline
newest firstAll Coverage
AI
Xiaomi Profit Falls 43% as Memory Chip Costs Bite
Xiaomi's adjusted net income fell 43% to 6.22 billion yuan in its third consecutive quarterly profit decline, as a persistent global memory-chip shortage -- driven substantially by AI data center demand for the same DRAM and NAND -- raised smartphone component costs faster than Xiaomi could pass them on.
DEEP DIVE
Xiaomi's HarnessX Rewrites Its Own AI Scaffolding Mid-Task -- and Smaller Models Gain the Most
AI
Xiaomi's HarnessX Rewrites Its Own AI Scaffolding Mid-Task -- and Smaller Models Gain the Most
Xiaomi unveiled HarnessX, a system that lets an AI agent rewrite its own scaffolding -- the prompts, tools and control logic around the model -- in the middle of a task. The standout finding: smaller, cheaper models benefit the most, suggesting clever orchestration can substitute for raw model size.