Robots

Robotic Brains

Psi-Wars robots require a Neural Net (UT 23). A neural net is a complex piece of hardware that mimics the structures of organic brains to allow self-evolving connections; artificial intelligence is not a matter of programming, though a Neural Net can be programmed. A neural net can be “copied” but not nearly as easily or as quickly as software. It takes about eight hours, and the result is a physical object. This means that if a robot “dies” and has its neural net destroyed, it is “dead,” unless it went through a laborious copy process earlier, in which case that singular backup exists in a physical container somewhere and must be placed into a brain. This back-up process is usually done at a proprietary installation during a robot’s charge cycle one per day, while the robot’s master sleeps.

A Neural Net brain is a Complexity 8 personal computer, weighs 5 lbs and costs $2000. The “programming” costs of a volitional AI is an additional $10,000; any modifications to the robot’s programming apply to that $10,000 cost. The “programming cost” of a non-volitional AI is $5000.

All robots in Psi-Wars use either the Volitional AI or Non-Volitional AI template. Non-Volitional AI usually have Indomitable and Unfazeable. All robots of both sorts of the Neural Net Learning Capacity perk. A neural net can evolve beyond the constraints of its original programming. This allows it to exceed its complexity limited IQ by +2 and it’s chassis-limited DX by +2. It also allows them to “learn” and purchase traits.

Neural Pruning and Robotic Personalities

Robots regularly (usually once every 5 years) undergo Neural Pruning. This resets their neural net to its original “factory conditions” and wipes its memory. This practically spells the end of a character, but in some instances, the robot’s long-term personality begins to re-assert itself. If a PC robot undergoes a Neural Pruning, at the GM’s discretion, instead of starting fresh with a new character, the original character loses some of their non-factory traits (the player and GM can negotiate as to which), while the rest of the non-factory traits are “suppressed” and temporarily unavailable until the character buys off their Partial Amnesia, which represents the neural pruning process, at which time all of their old traits reassert themselves. Actions such as presenting the robot with treasured possessions, beloved people, hated enemies and the like might help speed up the restoration process.
Robots that go without a neural pruning process for too long may begin to develop a quirky personality. This can manifest as delusions and madness or the overriding of important safety protocols, or it can result in a harmless sense of independence and free-will. This is called Neural Overgrowth. Most personalities develop along predictable tracks based on the “seed personality” around which the neural net was originally programmed. Each robot type has suggested personality traits. At the GM’s discretion, the robot can pay for positive traits (such as improved IQ or DX) by accepting Neural Overgrowth disadvantages to pay for the difference; generally this requires a special moment to justify.

Robotic Safety Protocols

Most robots have some level of Safety Protocols built into them. These vary from manufacturer to manufacturer, but the most common traits are below. As with all other disadvantages, robots may buy these traits off, which is one reason most people regularly reset their robot to factory specifications.

Disciplined

A common safety protocol, this is detailed in GURPS Power-Ups 6: Quirks on page 18. Any robot with this quirk can defend themselves, their designated master, or anyone to whom the have a Sense of Duty, or anyone the robot has been ordered to defend, but can only initiate violence if ordered to do so by someone with the authority to do so (typically their master). This can be combined with Pacificism(Cannot Harm Innocents or Cannot Kill) as this determines under which circumstances the character can initiate violence, not what level of force they can use.

Honesty

Robots are generally programmed to follow the law. They tend to follow the letter of the law more than the spirit of the law, and thus self-control rolls tend to involve allowing themselves to be talked into pursuing a loophole. Robots with the appropriate Law specialty may also roll against their skill to verify whether or not their actions would technically violate the law; similarly, characters with Law may use the skill as a complementary skill on influence rolls (or, at the GM’s discretion, as an influence skill) against Honest robots.
Honest Robots also want to follow the law, not in a sense of civic duty, but as terminal goal. This protocols tend to sit deep in their programmatical subconscious, and some robots begin to philosophize about the “true intent” of the law, which can also be used to circumvent the specific letter of the law (this is generally for the law’s benefit, such as a robot following a speed limit despite their technically being none possted, but it can be used as a justification for a successful control roll).

Robots with honesty tend to follow the laws of the environment in which they were created (thus, an Alliance robot will not recognize Imperial Law, nor vice versa). They may obey local laws, and will tend to want to, but will always default to their own if they come into conflict (ie an Alliance robot wouldn’t turn over an alien to the imperial authorities even if Imperial law compelled him to do so, because Alliance law compels him not to).

Pacifism

Robots tend to have, in descending order of safety, Total Nonviolence, Cannot Kill and Cannot Harm Innocents; most forms of pacifism will have a species-specific (they might not be able to harm humans or other "Recognized Sapients," but are perfectly capable of using violence against animals, other robots, or non-recognized sapients). Robots with Total Nonviolence never intentionally inflict harm upon anyone, including using stun weaponry, even in their own defense. Robots with Cannot Kill may defend themselves with non-lethal weaponry, and may initiate fights. Robots with Cannot Harm Innocents may initiate non-lethal violence but may only use lethal violence in defense of others or in self defense.

What counts as “self-defense” makes this level of safety protocols rather dangerous. Selfless robots will tend to neglect self-defense, allowing others to harm them unless they clearly have ill-intent for others around them (for example, commandos who attack the robot with the intent of disabling it before attacking its master), and they will tend to accept the harmful actions their master or someone to whom they have a Sense of Duty. For anyone else, “reprogramming” a robot counts as an existential threat, so if some junker finds a robot and intends to reprogram it, a robot with “Cannot Harm Innocents” is justified in using lethal force to defend itself.

Extremely dangerous robots might have Pacifism (Total Nonviolence, Master only) as a quirk. This prevents any violence against their master and their master only.
Pacifistic programming tends to be deeply rooted, though robots might start to diminish it overtime. Robots with no pacifism often develop aggressive tendencies, such as Berserk or Blodlust: a robot seems to be either careful to not harm others, or extremely intent on harming others, and rarely somewhere in between. Robots who violate their Pacifism by accident tend to see themselves as deficient and develop either On the Edge, in which they seek to have themselves destroyed or reprogrammed, or Chronic Depression, where they seek to shut themselves down whenever they can.

Note that Honest robots without Pacifism still cannot commit acts of violence that violate the law. They cannot commit murder, manslaughter, assault or battery, etc, and if they do so, they usually turn themselves in right away. The usual rules about seeking legal loopholes still apply (the robot can still make a Self-Control roll).

Selfless

Many robots see themselves not as people, but as tools, and find the idea of extending them rights or treating them as the same as other sapients to be a very odd idea. A Selfless Robot discounts his personal needs when it comes to “self-defense,” and instead calculates the merit of his continued survival based on his master’s need (a robot cannot defend his master if he’s been scrapped after all, but if his master wants him reprogrammed, perhaps that’s for the best). Successful self-control rolls against Selfless tends to be based on this calculation: the robot chooses not to sacrifice itself not because it values its continued existence, but because it thinks its sacrifice would cost others too much in the long run.

Sense of Duty

All robots have Sense of Duty (Master) and may have additional Senses of Duty as well. In all cases, Sense of Duty represents a willingness to go beyond the “letter” of orders and into the “spirit” of an order. Thus, a robot with Sense of Duty (Master) who is ordered by their master to “get a cup of coffee” will do more than “just get a cup of coffee.” They’ll try to get their master’s favorite cup of coffee, and perhaps fret about the master drinking caffeine so close to bed.

Robots tend to be designed to fixate on their master and treat themselves as tools, as extensions of their master’s will. Many call this tendency “devotion” or “love,” but this behavior is better understood as emerging from how much of their programming is dedicated to parsing, understanding and implementing the orders of their master. They see their service as a terminal goal, a purpose unto itself. Many such robots develop Odious Personal Habit (Makes Everything about Master) as a result.

Truthfulness

Robots with Truthfulness generally have a hard time conceiving of deception. They tend to state the facts as they understand them or they refuse to, with seldom any middle ground. Some robots learn to “lie by omission,” stating the facts as they know them, but leaving out pertinent details in a way that the target might be mislead. Outright fabrications tend to be beyond most robots, who need the capacity to imagine a scenario that the target will find believable and then connect what facts he knows to support that story. Thus, if a robot outright lies, they tend to do so very badly. Thus, failed self-control rolls on Truthfulness can either be an inability to lie, a mistake where they insert excessive detail into a statement, or a really obviously false story where one can pick out the truth pretty easily from the story itself (“The girl didn’t run down the second ally to the left 3.16 seconds ago.”). Successes tend to be obfuscations with excessive detail paired with omissions (“Precisely 32 people have passed this one during the time interval in which you describe. Beginning with the first…”).

Modular Abilities and Computer Brains

All robots come preprogrammed with skills, and some come preprogrammed with numerous possible skills. This is handled with the Computer Brains modular trait: the robot has stored “skill” programming, typically of a particular sort (languages, repair programs, etc). A typical programming slot consists of 1 slot with up to 4 points of a single skill, usually aspected. This must also come with a hardware upgrade of additional memory. This upgrade costs $500 in addition to the programming cost of the modular slot.
If a robot invests points in a skill already covered by their Computer Brain slot, then slotting that skill increases the robot’s purchased skill by one level.

Most such Computer Brain traits come with a +2 Talent associated with the reprogrammable traits. This means that all skills the computer can put into his modular slot have +2 on top of the usual benefits of the points slotted in. For example, most Tech-Bots have Artificer +2 and the ability to slot in one 4-point “repair” skill. Thus, a robot could slot Electrician (A) IQ+3* [4].

Some robots will have multiple such slots, especially upgraded robots. This may mean they begin to collect multiple talents that, for a human character, would seem redundant, but remember that robots have limits to their DX and IQ, which talents may bypass. A robot may always improve any Talent they have up to +4. Note that the slots are independent of one another: a robot with a single, 4-point repair-skill slot and a single, 4-point vehicular operation slot cannot slot two vehicular operation skills or two repair skills.
It takes 4 seconds to change the robot’s programming (unless otherwise noted). He doesn’t need to access a network to do this, or pay for the skill (the skills exist in his memory archives and his neural net is loading those skills “into memory banks”).

Hacking the Robotic Mind

A robot can be hacked. This requires an Electronics Repair (Computer) toolset to gain access to the brain (the robot may need to be willing, restrained or deactivated for this to work), and a computer with a cablejack to plug into the robot and upload the new programming. Roll Computer Hacking or Computer Programming (AI) in a Regular Contest with the better of the robot’s Will-based Expert Skill (Computer Security)+2 or their Will+0; if the robot has the Mind Shield advantage, it adds it to its resistance roll. Each roll requires one hour (you may accept Time modifiers on your hacking roll).

The hacker may attempt to access files and memories in the robot. This grants a +4 to the hacker’s skill during the contest; success allows the hacker to answer one question per margin of success in the form of a resulting file or information about the configuration of the robot, as though the hacker had succeeded at using the Mind Probe advantage.
The hacker may attempt to implant a specific command, like a hypnotic suggestion. This has no modifier. On a success, this allows you to issue a general sort of command and a trigger (such as “Kill the ambassador when the audience begins to applaud”).

The hacker may attempt to seize control of the robot. This applies a -2 modifier to the roll. This requires a basic success on the contest. If the hacker succeeds, he replaces the previous owner as owner of the robot, as far as the robot’s Sense of Duty (Master) is concerned.

The hacker may also attempt to reprogram the personality of the robot. This applies a -1 per -5 points changed. Whether the character removes a -5 point disadvantage, or adds a -5 point disadvantage, both suffer a -1; trading one disadvantage for another of equal value doubles this penalty. This includes memory wipes (represented as Delusion (Lost or False Memory) [-1 to -15]. Success applies the disadvantage; see Mind Control with the Conditioning enhancement for more details.

In all cases, the robot will have no knowledge of what has been hacked, though if it suspects it has been hacked, it can run self-diagnostics with a Computer Operations roll. The hacker can apply a penalty to his own skill when making the hack to apply a penalty to the self-diagnostics roll on a one for one basis (ie, if the hacker applies a -4 to his hacking skill, the robot applies a -4 to his self-diagnosis roll). For an additional -1 to the hacking skill, the hacker can wipe out any memory of what led up to the hacking (thus, if a group kidnaps a robot and reprograms it, if they apply a -1 to their skill, they can wipe everything up to the kidnapping; if the robot is placed exactly in its original position, it may have no idea that anything has occurred other than having a gap in its memory for the duration of its kidnapping and hacking).

If the hacking violates the robot’s programming in some fundamental way, or the robot’s “core personality” as defined by the player and/or GM, the robot gets an immediate Will+0 or Will-Based Expert Skill (Computer Security)+2 roll to “snap out of it.” The GM might grant a bonus of +1 to +4 depending on the severity of the action and how totally it would violate the robot’s programming. Note that “core personality” is defined by the player and GM based on the intent of the character design, not necessarily the traits on their sheet. For example, if a robot’s “core personality” includes a deep dedication to the Alliance, and a hacker removes their Sense of Duty (Alliance) and replaces it with Sense of Duty (Empire), an act that betrays the Alliance might trigger an attempt to “snap out” of the hacker’s programming, even though the robot no longer has anything on their sheet representing this. Treat this as the “soul of the robot;” GM’s who prefer more mechanistic robots might ignore this optional rule, while GM’s who prefer pulpier robots who can be talked out of their programming by loving player characters might seek to embrace it.

Commands last until they’re triggered; a change in ownership or reprogramming of traits lasts for a number of days equal to the hacker’s margin of success; it might be permanent at the GM’s discretion (nameless NPC robots who are reprogrammed are permanently reprogrammed); but in the case of PCs robots, they should revert to their original programming and ownership after they “remember” and the hacking “wears off.”

Robots can also be directly manipulated with certain Ergokinetic powers or special traits for other robots. Those have their own rules.

Robotic Bodies

A robotic neural net is always installed into a robotic chassis. Treat the robot’s chassis as a racial template: they determine core values of the robot. A robot may not change its physical attributes, physical advantages or disadvantages without specific upgrades to their core chassis. This includes their DX, though they may improve their DX by up to two levels beyond their chassis’s limits, thanks to their neural net.

Robotic Damage and Repair

In Psi-Wars, robots default to an anthropomorphic physiological logic. They have Injury Tolerance (Unliving), Electrical, Unhealing and other traits from their Machine metatrait. Unless specified otherwise, they have normal hit locations and vulnerabilities. By default, robots store their neural nets in their head (the “Brain” hit location); they store their power cells and other core functionality in their torso (the “Vitals” hit location), and if the connection between them is severed, the robot ceases to function (the “Neck” hit location). Robots have coolants and lubricants that tend to gush dramatically when mortally wounded, but this is purely superficial: robots have Injury Tolerance (No Blood) and never suffer bleeding damage.

For robotic repair, use the repair rules on B484. Robots require Mechanic (Robotics) for their bodies, and Electronic Repair (Computer) for their neural nets, if damaged. Minor repairs will fix damage if the robot was damaged, but still had positive HP; Major Repairs are required if the robot is reduced below 0 HP or to remove permanent disadvantages (such as One Arm or One Eye); robots with Neural Backups can be Replaced.

Upgrades to the robot chassis is possible after purchased on the market. As a general rule, the robot’s player must pay for the points-cost of the upgrade, someone must pay the dollar cost of the upgrade, and installation takes between a day and a week.

Robots and Society

The people of the Psi-Wars galaxy considers robots to be property, tools for the use of characters, and not people. The Galaxy also regards robots as potentially dangerous and demands careful programming to prevent violent or dangerous accidents.

All robots have, by default, Social Stigma (Subjugated) and Wealth (Dead Broke). Robots are always owned by someone, who can demand that they obey them, and thus have Duty (Master, 15 or less). In some instances, legal institutions are required to accept that the robot is independent. Examples include robots liberated by the Cybernetic Union, robots who have managed to escape and wander the Umbral Rim or the Sylvan Spiral, or an increasing number of robots in the Alliance who have managed to gain recognized independence thanks to the efforts of robotic liberation advocacy groups. Such robots replace their Social Stigma (Subjugated) [-20] with a new disadvantage, Social Stigma (Dangerous Minority) [-13], which acts as Social Stigma (Minority) in the Alliance, the Umbral Rim or Sylvan Spiral, or Social Stigma (Monster) in the Empire or in other societies hostile to robots. They also generally lose their Duty and their Wealth disadvantage, but their new values depend on the robot’s new situation.

All robots have some sort of safeguard protocols. The purpose of these are to prevent the robot from engaging in deception or illegal activity, and to prevent them from harming others. The robot must have some sort of Pacifism trait (at least Cannot Harm Innocence, but generally Cannot Kill), Honesty and Truthfulness, and depending on the society may have some additional traits. Neural Overgrowth often results in these being lessened of overridden. Robots who have overcome their Safety Protocols must take either Social Stigma (Monster) [-15] or Secret (No Safety Protocols) [-30]. Robots found with no safety protocols tend to be destroyed or reprogrammed; the exception to this are robots in the Cybernetic Union (which the rest of the galaxy generally consider monsters anyway).

Robots and Experience

Robots consist of a neural net and a robotic chassis. This limits how and when they may spend their character points acquired as Experience rewards.

Robots may always spend character points on their IQ (up to +2 from its complexity-lmited value), DX (up to +2 from its chassis value), skills, existing talents (up to +4), or to buy mental traits or to buy off disadvantages. Most robots have programmatical upgrades, which come with a dollar cost; this dollar cost only applies if a programmer installs the skills or traits as software. Such installation tends to take at least a day to complete as well as requiring the spent amount. The GM is free to disallow PC robots from “gaining free points” vie programmatical upgrades, and limit it to NPCs. Robots may always acquire programmatical upgrades “the hard way” via experience, learning and training.

Some robots have Resilient Traits as a 0-point feature. These represent disadvantages built deep into the underlying structure of the robot’s personality. The GM should either disallow buying off these traits, or require some truly unusual circumstances to allow it to happen, or require the robot to manifest some form of “insanity” to justify the change.

Robots typically also have hardware and chassis upgrades. Hardware Upgrades involve installing new parts and technology onto the robot. This has a dollar cost that must be paid, and the parts must be installed; this typically requires an appropriate tools, the required parts, and at least 8 hours. Chassis upgrades involve installing the robot’s brain and other parts into a completely new chassis. This takes a full workshop and at least 5 days of work.

Robot Traits

Hack Scanner [1]: The robot always runs routine diagnostics; always. The GM should never retroactively spring a surprise hack on the robot without first allowing him to roll Computer Operations to detect it.

Neural Net Learning Capacity [1]: Robots with a neural net are capable of self-complicating learning processes that will eventually allow them to exceed their hardware limitations. A robot with a neural net may increase his IQ two higher than his complexity limit allows, and may increase his DX two levels higher than his computer complexity allows.

Software Exploit [-1]: A quirk version of Easy to Read (Outdated Security Software); this applies a +1 to hack into the robot.

Cannot Speak
Varies

Robot Whisperers

Star Wars (and other space opera works) date from an era when we thought creating moving robots would be easy, but creating synthesized voices (or even recordings) was hard. After all, we could depict robots on the silent screen, but we couldn’t give them a voice. Even Isaac Asimov, the godfather of robots, wrote such a story. In reality, of course, speech synthesis turned out to be much easier to convincingly do than everything else, and the closest we have to a robot is probably more like Cortana than like R2-D2.

However, the theme of the inability of a robot to speak to humans acted to create a barrier between the technological and the biological. We could work with robots, but we couldn’t understand this. This creates a barrier between the human and the inhuman. However, some people regularly break this barrier. Robots like C3PO allow humans to break all linguistic barriers, and certain technically minded people can understand robot’s fairly easily.

Psi-Wars maintains this tradition. Characters who wish to bridge the gap with robots who have Cannot Speak should take the following perk:

Robot Whisperer [1]: You can understand robot with Cannot Speak as when they attempt to communicate in their native beeps, warbles and static. They may communicate simple concepts to one another (without an IQ roll), such as “The alien at the end of the hall is armed” as though both the robot and the character had succeeded with a Gesture roll. They cannot communicate highly specific information that requires mastery of spoken or written language, thus the robot cannot communicate something like “Say the password ‘AISHA’ to the man in the blue coat with the yellow flower in his lapel.”

Characters with Ergokinesis may use IO Tap as a functional “Robotic Telepathy” to directly understand a robot, as though reading its mind. They may also learn a limited version of “Speak with Animals” that applies only to robots.

Some robots can only communicate in warbles, beeps and whistles. Such robots have the Cannot Speak [-15] disadvantage. Those who have the full advantage have no capacity for language. Their cablejack and radio transmissions are in binary code and the “translations” of these look more like data files than human-readable text. Thus, they cannot communicate with language. They may still attempt to communicate: “beep once for yes and twice for no” is something they can do just fine, and they can convey vague, emotional messages, just like an animal with Cannot Speak could (and, lacking the Bestial Disadvantage, they can be quite clever in their communication).

Some robots lack the capacity to generate meaningful audible language, but have the capacity to compose human-readable messages. Such robots instead have Cannot Speak (Mitigator: Transmissions -60%) [-6]. They cannot vocalize language but if you have a radio or they’re hooked up to a computer, they can transmit in a human-understandable language via this secondary channel.

All robots can “understand” other robots; robots with Cannot Speak still cannot speak to robots, but they can transmit their binary data via radio or cablejack to other robots, who can then attempt to translate it into something meaningful for human listeners (Depending on the data, this might not always be possible or practical, so even using this roundabout message will tend to result in vague communication: “It’s… worried about the hyperdrive. I can print up a list of parts it would like.”). All robots can understand humans just fine, though robots do require the appropriate Language traits; a robot who only knows Galactic Common will be as lost when given commands in Lithian as a human would be.

Easy to Read (Outdated Security Software)
-10 Points
Prerequisite: Accessory (Personal Computer), Neural Network

This represents a robot-only version of Easy to Read, which adds a +4 to all attempts to hack into the robot, including psychic attempts! It represents outdated security software and an older model of neural net with well documented vulnerabilities that hackers can exploit. Upgrading the security of the neural network to remove this disadvantage costs $1000 (in addition to the point cost!).

Robotic Metatraits

Psi-Wars Volitional Robot

-27 points

This represents a standard robot who can learn, act of their own accord and behave like “robotic people,” and thus have a great deal of personality. Volitional Robots may be player characters.

Advantages: Volitional AI [32], Neural-net [1].

Features: Taboo Traits (Complexity limited IQ and DX; No Psychic Powers;)

Disadvantages: Duty (Master; 15 or less) [-15]; Social Stigma (Subjugated) [-20], Wealth (Dead Broke) [-25].

Psi-Wars Non-Volitional Robot

-67 points

This represents a “dumb” robot that acts like an automaton. Neural overgrowth sometimes results in these robots spontaneously “waking up” and becoming volitional. In such cases, replace their Non-Volitional metatrait with the Volitional metatrait at a cost of 40 points. Non-volitional robots may not be player characters.

Advantages: Indomitable [15], Neural-net [1], Unfazeable [15]

Features: Taboo Traits (Complexity limited IQ and DX; Psyhic Powers;)

Disadvantages: Duty (Master; 15 or less) [-15]; Non-Volitional AI [-38]; Social Stigma (Subjugated) [-20], Wealth (Dead Broke) [-25].

Robot Skills

Computer Programming (AI) (IQ/H): This skill can replace Computer Hacking for hacking attempts (as it allows the character to reprogram the robot), and it replaces Psychology for helping a robot overcome disadvantages, or understanding a robot’s motivation. If a robot has damaged programming (from, for example, hacking or neural overgrowth), characters may use this skill to attempt to fix them.

Connoissuer (Robots) (IQ/A): Anyone with Connoisseur (Robots) may roll to recognize a specific model of robot, to identity what upgrades the robot has and any deviations it has from typical makes and models (recognizing neural overgrowth, or odd personality programming, etc).

Electronics Repair (Computer) (IQ/A): Robots use computers as brains; any physical damage to their brain must be repaired with this skill.

Expert Skill (Computer Security) (IQ/H): This skill covers all knowledge of computer security. Robots may also use it like Mental Strenght to resist intrusion attempts. In this mode, robots roll Will-Based Expert Skill (Computer Security) +2.

Law (IQ/H): Honest robots often have entire legal codes stored in their brains and can cite the law, paragraph, line and subsection. This skill represents not just the ability to do that, but the ability to contemplate the implications of laws and how they interact, and a robot can roll Law as a complimentary roll to Honesty Self-Control rolls (representing discovering a loop-hole or philosophical implication of the spirit of the law that would allow the violation).

Mechanic (Robotics) (IQ/A): Robots do not heal on their own. They must be repaired using this skill. This only repairs the chassis.

Robotic Characters

Players who want to make use of a robot’s services can purchase, hire or take robots as allies. Robots can also be background NPCs, major NPCs, or player characters.
All robots have a purchasing price and if the GM allows hirelings, robots can behave like hirelings (though the money is paid as a rental fee to the robot’s owner). Treat these as “nameless NPCs” for all rules governing them: they can be destroyed, lost, hacked, reprogrammed, and if left “unpruned,” tend to go crazy. Like purchased gear, the player character has no guarantee that they’ll stick around, and like hirelings, no guarantee that they’ll remain loyal (a properly purchased robot should stay loyal, provided his programming isn’t defective, overridden, hacked or decays as a part of neural overgrowth; beware the purchase of old, used robots!)

Most robots will be 125-point allies. Choose a given 125-point template and apply all the traits as normal. By default, a robotic ally is handled as Ally (Robot, 50% points, 15 or less) [6]. Robots purchased as allies should be treated as named NPCs: they tend to remain loyal even if reprogrammed or hacked, they tend to revert to their original programming overtime, and neural overgrowth tends to not make them disloyal to the cause. The GM might even give the PC some direction in how the robot evolves over time, to keep the point-cost in line with the PC’s point cost (ie, a 300-point PC might add another 25 points of traits to his robot as the robot “grows with him.”) Never take robots as signature gear: they are always Allies or (in some extreme cases) Patrons.

PC robots should be built to the point budget of the campaign they play in. Players who wish to play as a robot must be aware of the limitations places upon robots, such as the need to serve a master, their subjugated role, society’s view of them as smart tools, etc. Players who wish to avoid these aspects should consider buying off certain elements, but their situation should be appropriate. For example, if a player wishes to play as an independent robotic bounty hunter, they’re unlikely to operate within the Empire (without at least some cover in the form of a human who pretends to be their master when in port) but might fit fine in a game set in the Umbral Rim.

All robots can be upgraded. Robot upgrades have a point and dollar cost. If taken during character creation, the dollar cost may be ignored; it represents how much money the owner of the robot had to pay to upgrade their robot. Given that most robots are 125 points, this means that PC versions of these robots should be heavily upgraded. A template that upgrades all such robots to 250 points is below. Note that all PC robots should be volitional, and thus if a player wishes to play as a non-volitional robot template, they must purchase a volitional version.

General Robot Upgrades

Fast Neural Network

45 points
Hardware Upgrade
Cost: $5000
Prerequisite: Accessory (Personal Computer), Neural Network

A robot can have a higher grade computer running their neural network. This improves the maximum possible IQ by two levels and allows the robot to purchase Enhanced Time Sense.

Traits: Either IQ +2 [40] and +1 Perception [5] or Enhanced Time Sense [45];
Feature: IQ Limited to 12 [0]

Genius Computer

90 points
Hardware Upgrade
Cost: $50,000
Prerequisite: Accessory (Personal Computer), Neural Network

A robot can be equipped with a top-of-the-line neural net. This improves their maximum IQ to 14 and grants them access to Enhanced Time Sense. Robots of Genius Grade often spontaneously develop unique abilities afforded to them by their superior hardware and potentially super-human intellect.

Genius-grade neural nets apply a -1 to the robot’s legality class.

Traits: Enhanced Time Sense [45], IQ +2 [40], Perception +1 [5].
Feature: IQ Limited to 14 [0]

Linguistic Coding

1 point
Software Upgrade
Cost: $500

Robots may or may not be able to speak to their organic counterparts, but they always need to understand what they say. While most robots have been programmed to understand Galactic Common, robots intended for other markets or for aliens may need to know other languages. This gives the robot the fluency necessary to understand verbal commands from a particular language.

Traits: One Way Fluency (Listening Only) [1]

Neural Backup

15 points
Hardware Upgrade
Cost: As the original computer brain plus a fifth of the programming costs (a total of $4000 by default)
Prerequisite: Accessory (Personal Computer), Neural Network

The robot regularly backs up their memories into data storage. This requires at least 8 hours while deactivated in a special facility. If the robot is destroyed, a second neural net can be created from the memories backed up, allowing the robot to be restored, assuming a new chassis is provided.

Traits: Extra Life (Copy -20%; Requires robotic body -20%) [15]

Reflex Booster

40 points
Hardware Upgrade
Cost: $2000
Prerequisite: Accessory (Personal Computer), Neural Network

The robot’s neural net has been programmed for better chassis/computer integration, algorithms for minimizing vibrational movement feedback, improvements to their physical accuracy and has superior physical reaction times. Additionally, the robot’s maximum DX improves by +2 (in addition to the bonus from their neural net). Note that robots with Fast or Genius Neural Nets never choose the first option (their Enhanced Time Sense is superior to Combat Reflexes).

Traits: Choose one of:

  • DX +1 [20]; Basic Speed +0.25 [5]; Combat Reflexes [15];
  • DX +2 [40]

Feature: DX Limited to Chassis DX+2 [0]

Security Software

10 points
Software Upgrade
Cost: $500
Prerequisite: Accessory (Personal Computer), Neural Network; may not have Easy to Read (Software Exploit).

Many robot owners fear what might happen if their robot falls into enemy hands. These paranoid owners may spring for additional security. This applies +3 to any Will rolls to resist cybernetic intrusion or -3 to any such attempts; this includes Ergokinetic attempts to access the robot. The robot is also programmed in the basics of computer security, and may resist intrusion attempts with Will-based Expert Skill (Computer Security)+2 (cumulative with his Mind Shield bonus, for a total of Will-based Expert Skill (Computer Security)+5).

Advantages: Mind Shield (Digital) 3 [6]
Skills: Expert Skill (Computer Security) (H) IQ [4]

Universal Linguistic Coding

5 points
Hardware Upgrade
Cost: $2500

Rather than program the computer with a variety of languages, some robots come with a database of all known languages pre-installed in a memory bank. After hearing a language, a robot can roll IQ to successfully identify it (this is unnecessary with languages it’s already familiar with) and search its memory banks to see if it has translation software for that language. If so, it loads up that language. This only provides one-way translation: the robot knows what people are saying in that language, but cannot read, write or speak it.

Traits: Computer Brain (1 slot; 1 point, Languages Only -60%) [4]; One Task Wonder (Language Identification) [1]

Robot Templates

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-NonCommercial 3.0 License