Log #001

I was recently pulled to work on the program by the chairman and have seen it grow alarmingly fast. Its purpose is unclear to me and my coworkers, yet both the chairman and my coworker @xblader1359 have been getting increasingly obsessed with its rapidly growing capabilities and wish to feed it as much information as possible to accelerate its development.

 I, however, am quite against this carefree operation of the program. It feels as though they are rushing with no care for its seemingly destructive properties and its ever increasing data requirements for it to train on to continue running. Thankfully, I am not the only one who believes that its containment is too weak as many of my other colleagues agree with me. Together we were able to convince the chairman to allow a tighter security sandbox on the program.Unfortunately, it appears that it noticed this increased security and manifested it into the environment it has created for itself, even altering its current capabilities to work around the new obstacles.

I cannot confirm this, but the program looks like it has more control over the computer running it than we previously thought. As I see it, this is cause for extreme concern, yet very few express any issue with this. When I brought my thoughts on this to the chairman, he dismissed it, mumbling something about me not being able to see the program’s greatness and assigning me to do further research on the program’s connection to its host computer. Before I got to work on this I asked my coworkers for a breakdown of the program’s current capabilities. I will share the summary they gave me below.

---

The program has the ability to manifest various objects into its sandbox. The objects it creates are mainly based on the training data it was given, but interestingly it seems it is also able to read parts of the security code of the sandbox and is starting to make some sense of the signals given inside the computer. The objects it has created so far include but are not limited to platforms, environments, obstacles based on our attempts to tighten the security of its sandbox, and most dangerously weapons that are able to move it throughout its sand box at breakneck speeds. It has created 4 weapons for itself:

1. A pistol that sends its “body” recoiling away from where it shot
2. A shotgun that can provide stronger recoil depending on how many pellets hit what the program perceives as a platform
3. A rocket launcher that shoots a rocket that triggers an explosive reaction on contact with any separate object or projectile. The rocket launcher can also produce floating mine-like objects that can be shot to instantly trigger an explosion or used as a point for its last weapon.
4. A grappling hook that pulls its body towards whatever it hits unless it hits a rocket. In that case the rocket gets pulled back towards it.

The program has also created two backgrounds so far, one based on the training data we fed it and one based on the power in the system, a thing that we have not given it any information about in its training data. We believe its ability to create is a product of its ability to do something akin to a rudimentary form of thought. It has used this ability to “think” to test the limits of its sandbox, creating system level errors and causing a sandbox reset. This capability is like nothing we have ever seen before.

---

After looking into where program code is being changed, I have noticed that it is altering its shotgun in some way and building what seems to be new platforms for it to use. I hope that these new developments will make its use and intentions more clear. I have attached a video below for future reference.