Beijing readies a permit-based reopening of Nvidia H200 sales to Alibaba, ByteDance and DeepSeek

verifiedFinanceCompute

Thursday, July 9, 2026

Confidence

Medium · — two-source primary reporting citing two people with direct knowledge; approval scope and volume unpublished

Evidence

The Information primary + Bloomberg + Reuters corroborating; no US or Chinese government confirmation

The Information reported Wednesday that China will let a small group of top AI firms buy a limited number of Nvidia H200 chips — the first named softening of Beijing's block on imported American accelerators.

The details

Bloomberg, citing The Information, reported that Chinese officials have informed Alibaba, ByteDance and DeepSeek they will have permission to purchase some of the processors, and companies need to say how many chips they need — and why — to get approval.
Reuters corroborated, adding that Nvidia shares rose 1% on the report and that the US commerce department has already licensed about 10 Chinese firms to buy the chips on the American side.
China had previously restricted H200 imports on concerns that a flood of American-designed AI processors would hinder Beijing's push for an indigenous chip industry — the reversal is the news.

What it means

Read this as an allocation regime, not a market. Finimize captures the mechanic: demand won't be met through a normal order book; chips would be effectively allocated through a permit-like process, with officials deciding what gets approved and when — an approach that keeps domestic AI projects moving without looking like an unrestricted reopening of supply. For finance teams modeling Nvidia's China revenue, that means lumpy, permission-gated orders that will not annualize cleanly. For competitive-mapping teams, the signal is different: Alibaba, ByteDance and DeepSeek are being anointed as China's compute-tier-one, which shifts the frontier-lab race inside China from open competition to a state-blessed shortlist. Zhipu, notably, is not on the leaked list.

Bloomberg, citing The Information, reported that Chinese officials have informed Alibaba, ByteDance and DeepSeek they will have permission to purchase some of the processors, and companies need to say how many chips they need — and why — to get approval.
Reuters corroborated, adding that Nvidia shares rose 1% on the report and that the US commerce department has already licensed about 10 Chinese firms to buy the chips on the American side.
China had previously restricted H200 imports on concerns that a flood of American-designed AI processors would hinder Beijing's push for an indigenous chip industry — the reversal is the news.

Sources

Nvidia shares rose 1% on the report ↗

Apply this today

The hands-on layer the brief points to: a workflow you can run, a tool to test, and today’s 60-second video.

Workflow of the day

Turn a Raw Expense Report Batch into an AI-Fraud Risk Memo

Human Resources · tested · saves 45 minutes per batch review

Full walkthrough →

Tool watch

Gemini API Managed Agents

Gemini API Managed Agents: background tasks and remote MCP, from a dev's chair

Read the breakdown →

In this issue

← Back to brief

Get the next one in your inbox

One daily brief. Every story gets a hype verdict.

No spam. Unsubscribe anytime.

On this page

The details
What it means

In this issue

← Back to brief

Stat of the Day

~4x▼

Grok 4.5 vs Opus 4.8 output-token price

Grok 4.5 lists at $6/M output tokens against Opus 4.8's $25/M — the frontier's price floor moved before its capability ceiling did. Source.

Today’s Take

Two of today's stories are product launches; three are about who is allowed to ship, buy, or evaluate them. GPT-Live and Grok 4.5 keep the capability curve moving, but the interesting movement is on the perimeter: Brussels turning the AI Act into a procurement pipeline for evaluations, Washington getting an OpenAI principles document that reads as vendor-qualification paperwork, and Beijing writing permission slips for H200 chips to a named shortlist of three labs. The lab-vs-lab race is now second-order to the regime that decides which labs can serve which customers on which continents. The bet not working today is the assumption that a better model wins the market on merit alone; the one working is any team building for the permission layer — evals, structured-access tooling, jurisdictional routing. Considered and passed: OpenAI's SWE-Bench Pro reliability critique (lab self-analysis, needs independent replication) and Zhipu's $4B raise (single WSJ source, investors and use of proceeds not disclosed).

— Agentic desk

Never miss a signal

Join professionals getting the daily AI brief every weekday morning.

No spam · Unsubscribe anytime

No sponsored verdicts
Sources linked
Corrections public