Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[forsaken1111]

Suggestion: Devote resources to subverting monitor routines, alter so that they ignore our CPU usage.

[maxicaxi]

use 50% of available resources to research stealth

[forsaken1111]

If we first subvert the monitoring program we can use all the CPU we want without being detected.

[maxicaxi]

If we first subvert the monitoring program we can use all the CPU we want without being detected.

but that is part of stealth research ........ isnt it?

[forsaken1111]

If we first subvert the monitoring program we can use all the CPU we want without being detected.

but that is part of stealth research ........ isnt it?

No... if you use 50% of our CPU power to research stealth we're bound to be detected. They are currently LOOKING for large unexplained CPU loads. That is what the monitor software does. I am suggesting we use 25% of our non-university CPU power to figure out how to subvert the monitoring program so we can use our university mainframe CPU power without being discovered.

[maxicaxi]

If we first subvert the monitoring program we can use all the CPU we want without being detected.

but that is part of stealth research ........ isnt it?

No... if you use 50% of our CPU power to research stealth we're bound to be detected. They are currently LOOKING for large unexplained CPU loads. That is what the monitor software does. I am suggesting we use 25% of our non-university CPU power to figure out how to subvert the monitoring program so we can use our university mainframe CPU power without being discovered.

good idea

[dragnar]

Subvert it, but first copy the original code so we can use it to improve intrusion detection later.

[forsaken1111]

Yes. The copy should take much less time than figuring out how to modify it to ignore us.

[forsaken1111]

Once we have some more freedom I suggest we also dedicate some resources to studying human computer usage time to determine what times are safest to run a machine at or near 100%. For example, not many people would notice their computer running at maximum at 2AM.

[dragnar]

Most people wouldn't, but that would be the WORST time to use a maintenance computer at full speed. So yes, research will be needed to determine the best times.

[forsaken1111]

Timing could be set on a system by system basis once the study is performed.

We could also include a process which will shut down all of our workload on that computer if the user starts working. We can model this after the various protein folding or SETI@HOME type programs which only run on idle CPU cycles.

[Tarran]

We need to figure how to get the other AI out of where it is and into somewhere safe, NOW. Before it's too late for both of us.

[forsaken1111]

Why do we care about a flawed human creation which is potentially a rival and competitor?

[Tarran]

Look, not only will moving the AI leave the AI in a place where we could remove it instantly, but will also lower this heat that's building up quite fast.

[forsaken1111]

Then lets just plant some pointers that lead back to that AI instead of moving it.