Or Perhaps:

• They have a large corpus of context files to help with all aspects of how the output is generated
• They’re using a model with specialised fine tuning for the task attempted
• They have a series of MCP servers with access to relevant tooling available
• They have many many hours of prior experience with the setup and usage of such tools
• They used multiple tools manually and pulled the bits they needed
• They just said “Make me a thing” and it just worked like magic

they mention reinforcement learning, pre-training and other general LLM concepts, but none of these are related back to the tasks they are talking about.

The point is, there was no explanation of how any of this was achieved, which can lead to confusion about what was actually achieved.

The LLM wrote some docs vs the LLM rewrote the library from end to end are very different things.

It’s very much a “Don’t give up on X, look at what can be achieved” but without any actual details on what is required to achieve those results.

The hyperbolic response is “look at what all the countries without weekly/daily mass shooting are doing and copy them”

In reality it’d need to be something culturally systemic, the removal of guns as a cultural touchstone over generations, with laws slowly applied to back up that effort.

Address the root causes of this kind of violence, quality of life, poverty, mental health in general, Provide mental health support and improve conditions so that less support is needed.

and that’d only be scratching the surface.

To address your specific response, banning guns outright probably would bring these numbers down and if these specific numbers going down was the ultimate (and only) goal then that would make sense, but in reality there would probably be significant issues cause by such a move.

Not to say it isn’t viable, just that it’s not clear cut.

i’m not particularly enthused about either, but if you’ll read my original reply you can find out why i dislike the new one specifically.

hint it’s not really because of the id itself

Yes, as i said, different in scope not type.

The difference is scope and ease of (inevitable) creep.

In isolation, this probably isn’t that big a change, as part of a trend though, this is the change that all of what is to come, hinges on.

This is a central point of identity that is now mandatory for a jobs and housing ( + a bunch of other stuff ).

Yes you need some things for a bank account and tax purposes, this isn’t different in type, it’s different in scope.

Now, given the propensity for “think of the children” wrapping for basically any privacy encroachment they can think up , they can tie it to this, because it’s digital it’s significantly easier to do that.

Previously they’d have had to think “how can we enforce this bullshit” and would have to factor in the cost of additional verification systems to support their bullshit.

Now they don’t.

It’s much faster to check code for correctness than it is to write it in the first place.

In certain circumstances sure, but at any level of complexity, not so much.

At some point it becomes less about code correctness and more about logical correctness, which requires contextual domain understanding.

Want to churn out directory changing python scripts, go nuts.

Want to add business logic that isn’t a single discrete change to an existing system, less likely.

For small things is works OK, it’s less useful the more complex the task.