Guozhen AIGlobal AI field notes and model intelligence

Realtime AI News

Blacklisted Chinese Firms Still Buying OpenAI and Google AI Tech via Singapore Loophole, Report Finds

An investigative report by Android Headlines reveals that Chinese companies on the U.S. Entity List are still acquiring AI technology from OpenAI and Google through intermediary channels in Singapore. Dubbed the Singapore Loophole, the practice raises questions about the effectiveness of existing export controls on AI technology.

Published
报道称被制裁中国公司仍通过新加坡渠道购入OpenAI和Google AI技术
Image source: gov.sg

An investigative report from Android Headlines has revealed that Chinese companies on the U.S. trade blacklist are continuing to access AI technology from OpenAI and Google through intermediate channels in Singapore. The report dubs this workaround the Singapore Loophole, suggesting that existing export control measures have a significant enforcement gap.

According to the report, these sanctioned Chinese companies procure and access U.S. AI technology and services through Singapore-based intermediaries or subsidiaries, circumventing the technology ban imposed on entities on the U.S. Entity List. Singapore's role as a global technology trading and financial hub provides a convenient environment for such transit operations.

The findings have reignited debate about the effectiveness of U.S. export controls on AI technology. In recent years, the U.S. government has steadily tightened restrictions on AI technology exports to China, covering high-performance chips, AI model weights, cloud services, and other domains. However, transit trade through third-party countries appears to have created openings in this regulatory framework.

Neither OpenAI nor Google has issued public responses to the specific claims in the report as of publication. The report does not disclose which specific Chinese companies are involved or the scale and value of the technology acquisitions.

The Singapore Loophole highlights a classic challenge for AI technology export controls: technology flows through globalized commercial networks, and unilateral restrictions often fail to close all possible circumvention routes. For policymakers trying to balance technological advantage with enforcement effectiveness, this loophole represents a critical regulatory blind spot.

If this channel persists, it could not only undermine the deterrent effect of current control measures but also incentivize more restricted entities to seek similar third-party transit paths, with significant implications for the global AI supply chain and geopolitical technology competition.

Why it matters

The exposure of the Singapore Loophole reveals a concrete enforcement gap in U.S. AI technology export controls, likely prompting increased regulatory scrutiny of third-party transit channels.

AI PolicyExport ControlChinaSingaporeOpenAIGoogleTrade
Back to realtime news

Nearby Updates

All

07/12, 01:39

OpenAI's GPT-5.6 Sol Ultra Solves 50-Year-Old Math Conjecture in Under an Hour

OpenAI announced that GPT-5.6 Sol Ultra has generated a complete proof of the Cycle Double Cover Conjecture, an open problem in graph theory that had remained unsolved for roughly 50 years. The model completed the task in under an hour using 64 parallel subagents, though mathematicians note the proof lacks proper citations of prior work.

07/12, 00:59

Microsoft Brings Its AI Agent Framework to Go After 8 Months of Development, Expanding Developer Reach

Microsoft has ported its AI agent framework to the Go programming language after approximately 8 months of development, enabling Go developers to build and deploy AI agent applications. The move extends Microsoft's AI toolchain to a language widely used in cloud-native and infrastructure development.

07/12, 02:38

Alibaba Open-Sources Open Code Review: An AI-Powered Review Tool for the AI Coding Era

Alibaba has open-sourced Open Code Review, an AI-powered code review tool that garnered over 5,000 GitHub stars in its first week. The tool uses a hybrid architecture combining deterministic engineering rules with AI agents to handle the growing review burden created by AI-generated code.

07/11, 22:52

Moonshot AI's K2.7 Code high-speed model officially lands on Kimi Code as a permanent mode

Moonshot AI's K2.7 Code high-speed model has been officially deployed on Kimi Code as a permanent selectable mode. The model focuses on faster code generation and programming assistance for developers.