Comment by cheriot
4 hours ago
Open-weight and OSS are wildly different and the article makes a poor comparison.
What's the incentive for the Chinese labs to continue releasing weights 5 years from now? It's not a stable equilibrium and cannot last.
- The lab spending large sums on research and training does not get the inference revenue to fund those efforts.
- Unlike OSS where a single volunteer can keep a project going, training costs run into the $billions.
- OSS is often a two way street where features and integrations are built that the original author benefits from. Open weight models are largely a one way street because the marginal benefit is so much less than training costs.
In the short term, it means Chinese labs can attract talent and, I suspect, funding from their gov. Similar to every other industry the CCP subsidized to take over.
China has no end of money to support these companies. The reason this equilibrium is unstable is that the autonomous agentic coding aspect of the models has been so successfully improved that it will soon be a threat to China state security.
> - Unlike OSS where a single volunteer can keep a project going, training costs run into the $billions.
Just some thought: Wouldn't it make sense to build some kind of volunteer computing project to train the next-generation LLM by volunteers, similar to the BOINC [1] projects or Folding@home [2]?
N.B.: BOINC was particularly famous for SETI@home (completed), Einstein@Home, Rosetta@home and PrimeGrid.
I still remember the time when Einstein@Home was in its heyday, and many people who loved putting together fast PCs contributed sometimes even for the reason of showing off in the statistics [3].
---
[1] https://en.wikipedia.org/wiki/Berkeley_Open_Infrastructure_f...
[2] https://en.wikipedia.org/wiki/Folding@home
[3] https://einsteinathome.org/de/community/stats
I’ll be impressed if somebody can make that work considering the vastly larger compute required.
> I’ll be impressed if somebody can make that work considering the vastly larger compute required.
I think you underestimate the computational ressources that the mentioned (and similar-kinded) scientific projects needed. Also consider how much computational ressources people invested into cryptocurrency mining.
No, I think the reasons are different:
- Many companies that train AI model use training data which must not be distributed for copyright reasons (and using it is a legal gray zone)x.
- Also consider that the amount of training data is insane. Scientific projects (and cryptocurrency mining, too) have the property that typically the amount of data (storage requirements) is small (or at least the computation can be partitioned so that each sub-task needs little data), but the required computing ressources are insane.
- AI companies consider a huge part of their training data as their "secret sauce" (they often even paid lots of money to generate it, for example by paying world-renowned experts for writing an answer for some important question).
Thus: Yes, the required computation ressources are huge, but this is a problem for which I consider it to be plausible that it can be solved. The real problems are in my opinion different.