One reasons people’s experiences differ here is that IP ranges of certain countries (outside of the very mainstream west) are much faster to get banned.
AI will give you only what you prompt it for, and the prompts are still designed by human analysts. What's changed with it is that now the 10K people surveilled will be anyone that "fits" in a category that the government deems criminal (say being transgender, or foreign, or left-wing, or criticizing ICE) instead of actual criminals. It shifts targeting, not capacity. If you wanted to be an actual criminal and are willing to do it in creative ways that bear no resemblance to other major categories of criminality (or lets be real, political disfavor), there is no time better than the present.
How do they counteract sham relationships/marriages. I assume they have to live together for a certain amount of time or something similar? If the marriage falls apart do they need to return the incentives?
Its funny - I was going to leave a similar comment - and I have, earlier, on a different thread. If people need fast classification, on a fairly scoped problem, it is very fruitful to start with an off-the-shelf embedding model like ModernBERT (which Laya uses) and stick a classifier in front - like a Support Vector Machine (SVM). For starters just tune the SVM, you don't even have to fine-tune the embedder - often it works very well, esp. given the compute needed. Plus you can get reliable confidence scores and generate explanations if you want them (using something like SHAP).
If you have an easy problem that can be solved by a classifier from pre LLM era, then sure, go ahead. But we have LLMs now and we can use those to expensively and slowly classify harder problems using general purpose models, without needing to spend a huge amount of time fine tuning a model for the specific task. Jev offers to do the same fast and cheap.
For clarity: no, a 0.8B model is not gonna do that.
I had the same thought: What were we all missing these past three years? And I think it's just that while everyone was staring at LLMs, the hardware has become a lot better and we now have 0.8B models that are extremely well-trained, so that allows us to look at the "old tech", such as classifiers, in a new light. It's also interesting because it decentralizes innovation: You don't need billions of hardware to do interesting experiments in this space.
The sun doesn’t set because the sun/earth moves, it sets because the earth spins. There’s no reason the setting sun should change which part of the moon is illuminated.
Claude code’s plan mode began its slow march towards deprecation when they hid the “clear context and implement” option behind a flag. From the discussion at that time it seemed that the developers considered it mostly a legacy feature, and that newer models are smart enough in long context not to need it.
This part I find a little disappointing. The plan is the perfect boundary to cut context which still isn’t free, even if models can handle a lot of it now.
This is why open harnesses and open models will always be superior, it's only a matter of time until someone decides that something you use isn't worth maintaining anymore. The reply from the Anthropic employee up above doesn't give me much confidence that Claude users will be able to continue using plan mode forever.
Codex and Claude Code are the best harnesses out there by a long shot. They work out of the box, require no real config, and just do the thing.
The flexibility and lightweight nature of open harnesses is mainly useful for open models. These models are considerably far behind the frontier and need a lot more steering. Complex harnesses confuse them. I use open harnesses for my OSS model rig because the model needs it.
> Codex and Claude Code are the best harnesses out there by a long shot. They work out of the box, require no real config, and just do the thing.
Even if I agreed, that doesn't address my point: Claude Code might be the world's best harness right now, but that doesn't stop an Anthropic employee coming along tomorrow and taking a giant steamy dump on the thing you like most about it. You're using their workflow, not your own.
Sure, but their workflow is the best for Claude, and it's the same with Codex for OpenAI's models. These companies work and live with their own harnesses, they are ahead of the curve on what works for their models and what doesn't.
reply