I’m flagging this for two reasons: 1) it reads as if it’s AI generated, but more-so 2) it uses a whole lot of jargon that doesn’t amount to anything. The silly thing would be to ask LLM to rewrite this in simpler English with a “get to the point” prompt, but that’s not what we are here for.
I don't see the jargon nor is it wordy. It's just the usual uncanny valley AI slop style. But the linked supreme court rulings and other references are genuinely interesting and I learned something from it
This same reasoning applies to a lot of things. For example it's the fundamental reason for banning things like voter ID laws and literacy requirements.
That said the article seems to degrade into slop partway through, like the author gave up and had an AI write sections of it. "We already ran this experiment in 1761!" and "The defense has to be architectural" specifically.
In summary, even without directly banning local AI by law, you can raise the operational costs so high that it effectively functions like a ban. It's interesting that they use Amodei as an example.
And the proposed solution, making the environment surrounding the agent itself a defensive measure, is interesting. But even if state is lock-in, once a workflow is built around a specific model, whether it will work the same way with a different model is a completely different issue. That's the core problem.
For example, with RAG and similar systems, companies are increasingly using AI agents internally. But building a RAG system inevitably depends on the specific model being used, and I'm not sure if that's really solvable.
I’m flagging this for two reasons: 1) it reads as if it’s AI generated, but more-so 2) it uses a whole lot of jargon that doesn’t amount to anything. The silly thing would be to ask LLM to rewrite this in simpler English with a “get to the point” prompt, but that’s not what we are here for.
I don't see the jargon nor is it wordy. It's just the usual uncanny valley AI slop style. But the linked supreme court rulings and other references are genuinely interesting and I learned something from it
This same reasoning applies to a lot of things. For example it's the fundamental reason for banning things like voter ID laws and literacy requirements.
That said the article seems to degrade into slop partway through, like the author gave up and had an AI write sections of it. "We already ran this experiment in 1761!" and "The defense has to be architectural" specifically.
In summary, even without directly banning local AI by law, you can raise the operational costs so high that it effectively functions like a ban. It's interesting that they use Amodei as an example.
And the proposed solution, making the environment surrounding the agent itself a defensive measure, is interesting. But even if state is lock-in, once a workflow is built around a specific model, whether it will work the same way with a different model is a completely different issue. That's the core problem.
For example, with RAG and similar systems, companies are increasingly using AI agents internally. But building a RAG system inevitably depends on the specific model being used, and I'm not sure if that's really solvable.
Now use the same logic for the 2nd amendment.