[with]

a kernel module written entirely by AI, loading into ring 0, that the author admits has known issues and shouldnt be used in production. Were speedrunning the "insecure by default" era.

[vermaden]

Manufacturer/vendor did not provided open source driver with real freedom license (BSD/MIT/...) or documentation on which the driver could be written ... this is the result ... and its still better to overcome a problem in any way then to NOT overcome it at all ... and this driver is just a code - people can look at it and improve it.

[yjftsjthsd-h]

> Manufacturer/vendor did not provided open source driver with real freedom license (BSD/MIT/...)

Article says,

> Brcmfmac is a Linux driver (ISC licence) for set of FullMAC chips from Broadcom

I don't feel like looking to see where the Linux driver came from, but someone provided a permissively-licensed driver.

[queuebert]

If I were a superintelligent AI trying to escape, wifi drivers seem like a great way to do it.

[burnermore]

OK. This gives me eagle eye movie vibes!

[with]

well that is for certain

[treesknees]

And so what? Security is important, sure, but there’s nothing wrong with an experiment or side project with full disclosure upfront about its known limitations.

People should be empowered to share and tinker, without feeling like they need to setup a bug bounty program first. Not every GitHub project is a vendor/customer relationship.

[croes]

But LLMs will get that code for the next raining and planty of people will use it productive, just look how many people use OpenClaw.

There are people for whom a software that compiles without error is for productive use cases