511–520 of 1001 posts
Distillation is fundamentally impossible to protect against. All you can do is slow them down. Change my view. Eventually these Chinese companies will release some extension like Honey, which will sit on top real, non-Chinese clients and send everything to China anyway. It's over.
Change my mind.
There's two basic kinds of distillation: 1) the massive [and dumb] method where you ask a question and use the answer as reinforcement (Black Box), and 2) more targeted distillation where you use one model to directly inform/train/guide another model (RLAIF). The latter is basically fine-tuning the model with direction from another model. Thousands of businesses do this every day to fine-tune. This is almost certainl…
> But if you show them a jailbreak of their model that bypasses their safety, they'll tell you that any model can eventually be jailbroken so don't worry about safety. They claim two things: 1) The specific, available jailbreak for Fable 5 is not dangerous - this has been confirmed by multiple experts, and there is no credible evidence against this claim (in other words, Anthropic is probably correct) 2) It is imposs…
Here's what is happening: Chinese resellers are offering Claude tokens at 70-90% below official Anthropic API prices. They achieve this by reselling capacity from pooled Claude Max accounts, payments fraud, and also reselling the model output & reasoning chains to various Chinese labs. They are subsidizing model access in exchange for user logs and reasoning traces, which they then sell as training data, allowing the…
Earlier quoted context omitted.
> This isn't "the market working as intended", this is an exhaustion fight to the bottom where the one with most money gets to stay in the market. As with most venture capital startups. I believe this VC tactic is a well documented "cheat code" to bypass market forces and build a monopoly. I find it hard to compare that with a free market. Why? Lots of people try this tactic, but hardly anyone ever succeeds. Meanwhil…
Lots of people have succeeded. Neither Anthropic nor OpenAI has any technical advantage in the field of subscription engineering.
Specifically, examples of people later exploiting their monopoly to charge people more than they otherwise would have paid.
> It said DeepSeek's operation involved over 150,000 exchanges
That volume seems more like the number of requests 15 employees using Claude Code would generate in a month. It seems too small for a large scale model distillation campaign.
Earlier quoted context omitted.
The AI companies seem to take the viewpoint that everything on the internet is free, except their stuff. It's okay to hammer some random website with AI crawlers, ignoring robots.txt, and causing bandwidth costs to skyrocket. But if you cost an AI provider money with your data acquisition practices, well, that's just clearly unacceptable.
That's one aspect, which is a bit of a gray zone. But Anthropic trained on pirated books. That is explicitly illegal.
There's two basic kinds of distillation: 1) the massive [and dumb] method where you ask a question and use the answer as reinforcement (Black Box), and 2) more targeted distillation where you use one model to directly inform/train/guide another model (RLAIF). The latter is basically fine-tuning the model with direction from another model. Thousands of businesses do this every day to fine-tune. This is almost certainl…
https://research.nvidia.com/labs/lpr/slm-agents/ - Distillation data is a natural byproduct of using these models. There's no effective defence against it. Anthropic is degrading thinking blocks to summaries to slow it down and hide model internals, but in the end, the math says you're SOL and it works on MNC/Large Corporate scale well enough that the moment cost becomes a priority, you're left without the lock in yo…
It’s the same reason why DRM for audio and video is a non sequitur - if you want a person to see or hear audio or video, eventually at the end of the chain, it’s going to be converted to audio for the ear and light for the eyes - that’s why you attach your tap.
Without a model generating tokens, what’s the point. So if Anthropic somehow disable quality token generation, what’s the point!
There's two basic kinds of distillation: 1) the massive [and dumb] method where you ask a question and use the answer as reinforcement (Black Box), and 2) more targeted distillation where you use one model to directly inform/train/guide another model (RLAIF). The latter is basically fine-tuning the model with direction from another model. Thousands of businesses do this every day to fine-tune. This is almost certainl…
The only way the U.S. keeps that edge is to prevent distillation. The only way Chinese companies can make up for the deficit in compute is to distill. There innovation in great supply on every side of the Ocean. Its about the chips. And in terms of national security, for the U.S., and for China, its about the chips and the distillation that undermines that advantage. This is an arms race.