Why are we restricting GPUs but selling Chinese labs our advanced data?
The US restricts the sale of high-end GPUs to China, treating them as strategic assets. However, American data companies that supply labs like OpenAI and Anthropic are reportedly also selling advanced training datasets to Chinese labs. This raises questions about the effectiveness of the GPU restrictions if the goal is to slow down competing labs, suggesting that advanced datasets might eventually face similar restrictions.
This report highlights a contradiction in US policy, as it restricts GPU sales to China but, unlike earlier efforts, allows advanced training data to be sold to Chinese labs.
Time & source
Times shown in UTC
Display time zone: UTC
Local time zone unavailable; showing UTC.
PublishedOffset at this time: UTC+0Sep 28, 2026, 23:05 UTC
IngestedOffset at this time: UTC+0Sep 29, 2026, 03:00 UTC
- Published
- Sep 28, 2026, 23:05
- Ingested
- Sep 29, 2026, 03:00
- Source type
- Dev community
- Tier
- Community
- Source status
- Healthy
Tier is a per-source editorial setting, not a per-item score.
Discussion trend
The percentage is based on collected discussion signal, not new comments or independent people. The curve only compares the same topic across time.
The US treats advanced GPUs like a strategic asset and still puts restrictions on high end chips going to China.
But at the same time, Forbes reported that American data companies supplying labs like openAI and anthropic are also selling advanced training datasets to Chinese labs.
This isn't basic labeling data you could scrape, it's expert written tasks, evals, coding data etc.
If the goal of the restrictions was slowing down competing labs then when will the advanced datasets share the same fate?