The BX-0 is not a product. It is a failed experiment that refuses to stop functioning. Lazarus's bioengineering division attempted to create a hybrid platform — mechanical chassis with biological tissue integration designed to give the automaton self-healing capability. The result was a quadruped platform roughly the size of a wolf, with a mechanical skeleton partially covered in lab-grown muscle tissue that contracts and relaxes in response to electrical stimulation from the internal control system. The biological tissue does regenerate, slowly, giving the BX-0 a limited ability to recover from damage that no purely mechanical platform possesses.

The failure is in the tissue's behavior. The lab-grown muscle responds to the control system's commands, but it also responds to stimuli the control system doesn't generate — it twitches, contracts asymmetrically, and occasionally drives the mechanical skeleton into movements that weren't commanded. The BX-0 stumbles, jerks, and lurches in ways that are profoundly disturbing to observe. It moves like something in pain, though it has no capacity for pain. The biological tissue occasionally grows in uncontrolled patterns, producing tumor-like masses that must be surgically removed before they interfere with joint articulation.

Lazarus has produced twelve BX-0 units, each slightly different as the bioengineering team iterates on the tissue integration process. None of them work correctly. None of them have been decommissioned. They are kept in a sealed laboratory section, powered and active, while researchers continue to study their failures. Lab personnel who work near the BX-0 containment area report persistent unease that they attribute to the sounds — the wet mechanical grinding of biological tissue being driven by electrical impulse against metal bone.