Commerce Department Backlog Stalls Nvidia H20 Chip Licenses

Why Nvidia H20 chip licenses are being stalled

If you've been wondering why Nvidia H20 chip licenses haven’t been approved yet, you're not alone. In July 2025, U.S. Commerce Secretary Howard Lutnick announced that Nvidia could once again apply to sell its H20 AI chips to China. Yet, despite that green light, Nvidia still hasn't received an approved export license. This post explains why the delay is happening, what’s at stake for both U.S. firms and global AI competition, and how the ongoing backlog reflects deeper bureaucratic turmoil. Read on to get a clear answer to the top questions users search for—Why are the licenses delayed? What does it mean for Nvidia and China? And what might change next?

Image Credits:Li Hongbo/VCG / Getty Images

**1. Why Nvidia H20 chip licenses are delayed

The Commerce Department’s Bureau of Industry and Security (BIS) is currently facing a historic backlog of export license applications. According to Reuters, thousands of applications—including those for Nvidia’s H20 chip—remain unprocessed due to widespread staffing losses, a breakdown in communications, and internal micromanagement under Undersecretary Jeffrey Kessler.  The backlog is reportedly the longest the agency has seen in over three decades. BIS averaged 38 days per license in fiscal 2023, denying only about 2% of applications; today, the delays stretch much longer—and distributions remain frozen. These delays aren’t just technical hiccups—they’re symptomatic of real operational chaos affecting U.S. export policy.

**2. Why Nvidia H20 chip licenses matter

These Nvidia H20 chip licenses are pivotal for several reasons. First, Nvidia h ad hoped to resume H20 shipments to China shortly after the Biden‑era ban was reversed under the Trump administration in July 2025. Instead, the lack of actual license approvals has stalled deliveries worth billions of dollars—Nvidia had already ordered an extra 300,000 H20 chips from TSMC to meet surging demand from firms like Tencent, Alibaba, and ByteDance. Second, national security experts and U.S. lawmakers have warned that exporting H20 chips could accelerate China's military and AI capabilities  Nvidia insists its chips have no backdoors and are compliant with U.S. controls, but China’s own regulator recently summoned the company over security concerns.The presence—or absence—of those licenses may well shape the balance of AI power in the global tech race.

**3. What the backlog reveals about U.S. export policy

The Nvidia H20 chip licenses delay underscores deeper structural issues in U.S. export control management. Sources describe micromanagement by BIS leadership, which has hindered responsiveness, slowed rulemaking, and cut off meaningful communication with industry.  Key regulatory updates promised months ago have yet to be published, including revisions to a Biden‑era rule restricting AI chip exports.  Meanwhile, experienced staff have resigned or left through buyouts, leaving critical roles vacant—even at top enforcement offices.  Trade and compliance experts warn that ongoing unpredictability is damaging U.S. competitiveness and incentivizing partners to look elsewhere. 

**4. What comes next for Nvidia H20 chip licenses—and the AI race

Looking ahead, resolving the Nvidia H20 chip licenses backlog depends on restoring operational discipline at the Commerce Department. That may require clearer policy guidance, reinvigorated staffing, and renewed engagement between regulators and industry. If licensing resumes, Nvidia may begin recognizing revenue from China-bound H20 shipments by fall 2025 or early 2026. Analysts estimate China demand could exceed 1.8 million units, far beyond Nvidia’s existing inventory of 600,000–700,000 chips.  But delays could give domestic Chinese chipmakers more time to catch up, eroding Nvidia’s market share faster. On the geopolitical front, trade talks—tied to rare earth negotiations—may further influence how quickly licenses are issued.  How swiftly BIS acts over the coming weeks will signal whether the U.S. can balance national security with global AI leadership.

Post a Comment

Previous Post Next Post