It's now spit out about 40 tokens after maybe 18 hours and has not finished the "thinking" stage of responding to the prompt. I'll let it keep running to see what happens

The USB drive light is flickering, showing something is happening. It's been about 8 hours since I entered the prompt and I've gotten about 10 tokens back so far. I'm going to leave it running overnight and see what happens.

Instead of "programs that do one thing and do it well", "write programs which are designed to be used together" and "write programs to handle text streams", I might go with a foundational philosophy like "write programs that are do not trust the user or the admin" because in applications connected to the internet, both groups often make mistakes or are malicious. Also something like "write programs that are strict on which inputs they accept" because a lot of input is malicious.

Unfortunately I think things are moving so fast that by the time such a study was done, we would already be on to newer models and newer versions of gas town.

I tried tracking down where those numbers came from and the sources were a bit sketchy. Can anybody who has used Gas Town confirm those numbers, or report their personal numbers?

Also, in addition to underground and outer space, we should consider underwater. Underwater bases would be safe against most missiles and drones. Cargo submarines could bring gear to our bases safe from drones and anti ship missiles. And we may want to revisit the idea of a submarine aircraft carrier but with drones instead of manned aircraft.

Seems to me that one drastic tactic NPM could employ to prevent attacks like this is to use hardware security. NPM could procure and configure laptops with identity rooted in the laptop TPM instead of 2FA. Configure the NPM servers so that for certain repos only updates signed with the private key in the laptop TPM can be pushed to NPM. Each high profile repo would have certain laptops that can upload for that repo. Set up the laptop with a minimal version of Linux with just the command line tools to upload to NPM, not even a browser or desktop environment. Give those laptops to maintainers of high profile repos for free to use for updates.

Then at update time, the maintainer just transfers the code from their dev machine to the secure laptop via USB drive or CD and pushes to NPM from the special laptop.

As far as humans in the loop, the only human we ultimately cannot get rid of is the user. But I think with a combo of user feedback forms and automated metrics we can give AI a lot of feedback about how good software is just from users using the software.

Like, can an AI agent use a browser, attempt to use the software, find bugs and create a ticket? Can an AI agent use a browser, try to use the software and suggest new features?

Feels to me like there are at least one or two more paradigm shifts coming in how AI gets used which will make current tools obsolete.

As one example, I think we will eventually get GUI dashboards to manage AI agents which will be easier to use than current CLI tools.