Override All Security! Computers and Hacking in Starfinder

Gotta have computer skills in Starfinder. I've already got some ideas that work off the "Artificial Personality" thing...

Make Computer Use a simple INT check, DC = 10 + 2 X Computer's Complexity. Computer Programming / Hacking should be actual skills. Just like today. Nearly everyone knows how to use a computer. But actually programming or hacking one takes education.

Edit: Oops, posted before I read the above text. Feel free to ignore.

So let's just put it out there: Like many RPGs set in a universe with advanced technology, Starfinder has rules for computers, and hacking computers. And, like nearly all RPGs that do so, these rules aren't grounded in reality.

Thank you for heading that discussion off at the pass.

Beyond its tier, a computer is defined by its size, user interface, access and authorization, basic function, modules (which define what the computer can do or control beyond its basic functions, and may include controlling other devices or computers), and countermeasures.

There is a ton of depth to unpack here. Using the Computers skill is so much more than just making a single roll - it's a form of micro combat, where you need to choose what modules or functions to take on, especially if time is an issue. You could hack ten computers in a day and each one could be a unique encounter with its own challenges.

Also, there isn't a lot that a player needs to know in order to navigate those ten different computers; all she needs to do is make a Computers check. All the GM needs to know is how the DC scales for different tiers and subsystems. Obviously, this is a brief overview, but it looks like it will be a good balance between depth and simplicity. I hope most of the major skills look like this.

Make Computer Use a simple INT check, DC = 10 + 2 X Computer's Complexity. Computer Programming / Hacking should be actual skills. Just like today. Nearly everyone knows how to use a computer. But actually programming or hacking one takes education.

Edit: Oops, posted before I read the above text. Feel free to ignore.

As far as we can tell from what was shown so far, Starfinder made streamlining and consolidating skills a big point, which I appreciate a lot.

I appreciate that physical presence looks to be required for most hacking - keep the hacker with the group, dang it!

blog wrote:

So let's just put it out there: Like many RPGs set in a universe with advanced technology, Starfinder has rules for computers, and hacking computers. And, like nearly all RPGs that do so, these rules aren't grounded in reality.

Thank you for heading that discussion off at the pass.

Yeah, no kidding. Moreover, the genre conventions by and large wouldn't support realistic hacking anyway - we want R2 plugging into one of those little rotational sockets and gaining access to prisoner manifests and engineering schematics in like 30 seconds.

Greetings Gentle Sentient,

My name is Prince PugWumpus IV, and I hope this e-communication finds you in good health. Recently, I and my family befell some false and libelous legalistic circumstances on Castrovel. While my family was able to escape off-planet, I have remained behind to help ensure our vast fortune is safely transferred to the First Bank of AbadarCorp on Absalom Station. In order to avoid the attentions of the tenacious scoundrels slandering my family, I need to transfer our trillions of latinum quatloos through a complex and circuitous route... and this is where you come in. If we could quickly and quietly route our funds through your personal e-Abadar account, I would reward you handsomely with 10% of our fortune. If you could simply forward me the details of your e-account, I c- ARRRRRRRRGH

<BLACK ICE: SUCCESSFUL; SENDER: OFFLINE>
<Thank you for using Epoch Systems E-Security>

While this isn't the most exciting blog for me, it does wet my appetite for more.

I like these rules a lot. Anything you can tell us about infospheres, and how computer skills interact with them? Are there any matrix-esque systems?

I like these rules a lot. Anything you can tell us about infospheres, and how computer skills interact with them? Are there any matrix-esque systems?

While I presume a lot more detail will exist on both of those kinds of things in time, we don't take up a lot of space for them in the core rules.

Do the rules allow for the classic trope of smashing the control panel to get the result you want? For that matter, Can AI's be threatened with destruction to get what you want?

Oh! Do Starfinder robots follow the 3 laws?

There's no mention of any 3 laws that robots are assumed to follow.

Robots can be deadly as frak.

So I'm assuming somewhere out in the recesses of space, there might...SKYNET?!!

Moreover, the genre conventions by and large wouldn't support realistic hacking anyway - we want R2 plugging into one of those little rotational sockets and gaining access to prisoner manifests and engineering schematics in like 30 seconds.

Which makes me wonder if the Engineers Drone can do such things ala R2D2/Ghost(from Destiny)/Chopper.

'You must guard the hacker to give him enough time to bring down security!'

So I'm assuming somewhere out in the recesses of space, there might...SKYNET?!!

Kirk's Enterprise ran into three or four different variations over the years, as I recall.

I'm looking forward to seeing how these rules work in-game.

As I'm heavily influenced by Peter F. Hamilton's work, at the moment, I'm interested in seeing how I can work up things like his setting's bitek, in which living ships (voidhawks, blackhawks, and the living habitats) have fully developed consciousness of their own.

Of course, I'm also hooked on the idea of nanonics as a means of directly interfacing computer systems by thought.

This video demonstrates the Starfinder computer skill being used.

Will There be any rules for true Artificial Intelligences in the core book?

I'm perfectly fine with the "hacker needs physical access" trope, because it is plausible enough - in the real world, the computers with the REALLY good stuff are still air-gapped and WI-FI shielded. (Or, let me correct myself: if PEOPLE WERE DOING THEIR ****ING JOBS CORRECTLY, the good stuff is still air-gapped and Wi Fi shielded. :)

Will There be any rules for true Artificial Intelligences in the core book?

I'd like to know this as well. The AI upgrade up there doesn't seem to do anything other than let your computer fake a conversation.

There's no mention of any 3 laws that robots are assumed to follow.

Robots can be deadly as frak.

Security Bot: "Asimov's Three WHAT, now?"

Sure. androids are true AIs. That's a core race.
The mechanic has a class feature called "Artificial Intelligence"
We mention there are robots on Aballon that are self-aware.

But no, the equipment section does not let you buy or sell self-aware creatures, for what I think are fairly obvious reasons.

Brew Bird wrote:

Will There be any rules for true Artificial Intelligences in the core book?

I'd like to know this as well. The AI upgrade up there doesn't seem to do anything other than let your computer fake a conversation.

There's a reason that ability is called "Artificial Personality."

I noticed a distinct like of "is there a Skynet?" in your answer Owen.. ;)

Can AI's be threatened with destruction to get what you want?

Or instead of destruction, maybe to inhibit the AP/AI from doing it's primary or other functions.

Does this have a definite answer, or will this be left up to GMs to choose to implement?

Computers skill checks have DCs based on the tier and countermeasures of the computer they are applied to, and generally take one round per tier of the computer.

A tier 5 system takes 30 seconds to breach. That sounds quick, but then you realize you need to hold off the security drones for six rounds without your technomancer.

Sure. androids are true AIs. That's a core race.

The mechanic has a class feature called "Artificial Intelligence"
We mention there are robots on Aballon that are self-aware.

But no, the equipment section does not let you buy or sell self-aware creatures, for what I think are fairly obvious reasons.

You know, I never stopped and considered the implications of putting a price on intelligent magic items until just now.

Of course I totally understand why the equipment section would omit buying or selling self-aware creatures. Tangentially, though, does this mean that intelligent magic items will no longer be a thing, or no longer be a thing with a system value placed on them?

As a thought experiment (I do absolutely agree it would be a bad idea to publish the ability to buy and sell self-aware creatures anyplace that isn't the Evil League of Evil Handbook), wouldn't the cost in credits simply be equivalent to the requisite number of UPBs required to create an AI? Or is there some kind of Full Metal Alchemist-esque limitation where you can't simply science a being with a soul into existence?

... Though I suppose Pathfinder does set the precedent for spells that awaken constructs and animals... and if it's a relatively standard practice to sell spellcasting services for credits, the same way spellcasting services could be exchanged for gold, one would just add the cost of hiring someone to cast the spell to the AI...

Sure. androids are true AIs. That's a core race.

The mechanic has a class feature called "Artificial Intelligence"
We mention there are robots on Aballon that are self-aware.

But no, the equipment section does not let you buy or sell self-aware creatures, for what I think are fairly obvious reasons.

By rules for AI, I meant more like the AI in the technology guide. How an artificial intelligence works as a not-quite creature, without a physical body. What they are and are not capable of, and how adventurers can disable them, etc. Granted, that's probably something that belongs in the Alien Archive, but is it something we can expect to find somewhere in Starfinder?

Such computers are often given a name and are capable of parsing expressions, slang, social cues, tone of voice, and similar elements beyond a literal understanding of spoken or written words.

...
developed by programmers over centuries to allow for extremely natural-sounding conversations.

So if players found a really old ship adrift in space or they couldn't buy a model made this century ect and the computer was old would for example the Artificial Personality (if it had one) become less socially aware or is it something that has been that way for hundreds of years now?

I would say that any A.P. that was out of the cultural development loop for a while would remain old fashioned until its algorithms could pick up new terminology. If its programming allowed it to do so. It might have had a particularly fussy owner that required a strict adherence to a particular etiquette. For instance, a military craft might have an A.P. that was incapable of anything less than a simulated military bearing.

random roll
your data pad yawns

I would say that any A.P. that was out of the cultural development loop for a while would remain old fashioned until its algorithms could pick up new terminology. If its programming allowed it to do so. It might have had a particularly fussy owner that required a strict adherence to a particular etiquette. For instance, a military craft might have an A.P. that was incapable of anything less than a simulated military bearing.

You say old fashioned... And for some reason my brain thinks British.

Or at least, industrial era British as heard by an American in the 1990s... If the 1990s was what the 1990s thought the 2090s would be. So basically C3PO, or Alfred (batman's butler), or Jarvis. And of course, such a professional, almost corporate speak way of talking would make the personality profile sound less artificial if the code's still incomplete.

Let's all be honest though, I think we're more interested in what the slang in Starfinder would be rather than what the personality code is like. Given that if we knew what the slang 100 years from now is going to be, we'd probably sound a lot like whatever the personality thing from "a very old wrecked ship" sounds like.

Meanwhile, I notice a certain point of interest: if the DC is potentially equal to your bonus, and you need to beat the DC by 20 to control the root system, then one plan may be to have one PC try and take 20 on the computer check, while the rest of the team buys time to gain control of the rest of the systems. That said... That would take a lot of time, and the only reason you'd be doing that when infiltrating a facility, is because you figure you need to basically own the place. All access pass and whatnot.

That said... Such a scenario is probably unheard of. I find it highly unlikely that a well balanced AP would have Mission Critical Computers like that, being somehow so easy to hack that a given system is able to be hacked on a Nat 1.

The "beat DC by 20 to stop making checks" feels very superfluous and pointless to be honest.
To be even theoretically able to beat DC by 20, you need to be able to pass the DC at a roll of 1.
And, at that point, you autosucceed every check anyway, so you wouldn't likely have to roll in the first place.

The "beat DC by 20 to stop making checks" feels very superfluous and pointless to be honest.

To be even theoretically able to beat DC by 20, you need to be able to pass the DC at a roll of 1.
And, at that point, you autosucceed every check anyway, so you wouldn't likely have to roll in the first place.

Don't forget one-time bonuses and the like, though, not to mention the possibility of special effects that lower this threshold. In most cases, you're probably right, but this does leave room open for a lucky roll using limited resources to make things even easier on you later.

Mashallah wrote:

The "beat DC by 20 to stop making checks" feels very superfluous and pointless to be honest.

To be even theoretically able to beat DC by 20, you need to be able to pass the DC at a roll of 1.
And, at that point, you autosucceed every check anyway, so you wouldn't likely have to roll in the first place.

Don't forget one-time bonuses and the like, though, not to mention the possibility of special effects that lower this threshold. In most cases, you're probably right, but this does leave room open for a lucky roll using limited resources to make things even easier on you later.

Funny how "make things easier on you later" basically amounts to giving you the keys to straight up turn the whole complex the computer runs against its owner.

Assuming its the computer running the BBEG's base; At that point, you've hit the giant "I win" button on anything that's not a complete boss, or a counter-hacking operative. Assuming the computer isn't a glorified library that is. You can basically lock down all doors, set the turrets to fire at anything, and erase the tapes.

Then again, I'm probably not giving security experts the proper credit. They aren't likely to purposefully have ALL systems running off the same computer, or all sectors of a base doing so. That's like ASKING to be hacked. A BBEG's base should in theory have at least three separate computer systems for the turrets, another for physical access doors and personnel access, two separate grade computers for high and low grade information, and 2-4 counter-hacking operatives. Meanwhile, the "Rootfile" should be reconfigured to reset any possible "pre-meditated" intrusion attempts, AT LEAST once a week. Twice a week if they're cautious.

If the computer is archiving information or security that COULD change the fate of an entire planet... It also needs to be guarded by one very devoted "True AI". Preferably one created by Triune themselves if possible, like some kind of child gifted by the gods like something out of Wonder Woman.

Funny how "make things easier on you later" basically amounts to giving you the keys to straight up turn the whole complex the computer runs against its owner.

Assuming its the computer running the BBEG's base; At that point, you've hit the giant "I win" button on anything that's not a complete boss, or a counter-hacking operative. Assuming the computer isn't a glorified library that is. You can basically lock down all doors, set the turrets to fire at anything, and erase the tapes.

Then again, I'm probably not giving security experts the proper credit. They aren't likely to purposefully have ALL systems running off the same computer, or all sectors of a base doing so. That's like ASKING to be hacked. A BBEG's base should in theory have at least three separate computer systems for the turrets, another for physical access doors and personnel access, two separate grade computers for high and low grade information, and 2-4 counter-hacking operatives. Meanwhile, the "Rootfile" should be reconfigured to reset any possible "pre-meditated" intrusion attempts, AT LEAST once a week. Twice a week if they're cautious.

If the computer is archiving information or security that COULD change the fate of an entire planet... It also needs to be guarded by one very devoted "True AI". Preferably one created by Triune themselves if possible, like some kind of child gifted by the gods...

Then again there is this bit in the above section.

So let's just put it out there: Like many RPGs set in a universe with advanced technology, Starfinder has rules for computers, and hacking computers. And, like nearly all RPGs that do so, these rules aren't grounded in reality.

:)

Shisumo wrote:

Moreover, the genre conventions by and large wouldn't support realistic hacking anyway - we want R2 plugging into one of those little rotational sockets and gaining access to prisoner manifests and engineering schematics in like 30 seconds.

Which makes me wonder if the Engineers Drone can do such things ala R2D2/Ghost(from Destiny)/Chopper.

'You must guard the hacker to give him enough time to bring down security!'

I would be deeply saddened if this wasn't possible!

Sure. androids are true AIs. That's a core race.

The mechanic has a class feature called "Artificial Intelligence"
We mention there are robots on Aballon that are self-aware.

Hmm, I'm going to go ahead and assume that this applies to the drone/exocortex, which is very exciting!

Sure. androids are true AIs. That's a core race.

The mechanic has a class feature called "Artificial Intelligence"
We mention there are robots on Aballon that are self-aware.

But no, the equipment section does not let you buy or sell self-aware creatures, for what I think are fairly obvious reasons.

I thought androids explicitly aren't AI in the first place?

Owen K. C. Stephens wrote:

Sure. androids are true AIs. That's a core race.

The mechanic has a class feature called "Artificial Intelligence"
We mention there are robots on Aballon that are self-aware.

But no, the equipment section does not let you buy or sell self-aware creatures, for what I think are fairly obvious reasons.

I thought androids explicitly aren't AI in the first place?

Instead, they get normal souls granting them sapience just like any other living creature.

I believe in one of the Paizocon panels it was stated that the line where a sufficiently advanced computer becomes a true artificial intelligence is when it gains a soul.

I don't have the reference for that unfortunately.

Which makes me wonder if the Engineers Drone can do such things ala R2D2/Ghost(from Destiny)/Chopper.

'You must guard the hacker to give him enough time to bring down security!'

This makes me wonder if, using the action split, you yourself can guard your drone while it hacks or something. Or. if you can split your actions enough, so you are hacking and its guarding you.

That was a thing that always bothered me about Shadowrun's Harebrained scheme games. You couldn't Deck and have your drones protect you. Split attention I suppose... but in this... you're drone will have at least a supplimentary AI to it.

I'm looking forward to seeing how these rules work in-game.

As I'm heavily influenced by Peter F. Hamilton's work, at the moment, I'm interested in seeing how I can work up things like his setting's bitek, in which living ships (voidhawks, blackhawks, and the living habitats) have fully developed consciousness of their own.

Of course, I'm also hooked on the idea of nanonics as a means of directly interfacing computer systems by thought.

So much this^ I love Hamilton's work and would love to see that in the setting.

Luna Protege wrote:

Funny how "make things easier on you later" basically amounts to giving you the keys to straight up turn the whole complex the computer runs against its owner.

Assuming its the computer running the BBEG's base; At that point, you've hit the giant "I win" button on anything that's not a complete boss, or a counter-hacking operative. Assuming the computer isn't a glorified library that is. You can basically lock down all doors, set the turrets to fire at anything, and erase the tapes.

Then again, I'm probably not giving security experts the proper credit. They aren't likely to purposefully have ALL systems running off the same computer, or all sectors of a base doing so. That's like ASKING to be hacked. A BBEG's base should in theory have at least three separate computer systems for the turrets, another for physical access doors and personnel access, two separate grade computers for high and low grade information, and 2-4 counter-hacking operatives. Meanwhile, the "Rootfile" should be reconfigured to reset any possible "pre-meditated" intrusion attempts, AT LEAST once a week. Twice a week if they're cautious.

If the computer is archiving information or security that COULD change the fate of an entire planet... It also needs to be guarded by one very devoted "True AI". Preferably one created by Triune themselves if possible, like some kind of child gifted by the gods...

Then again there is this bit in the above section.

Owen KC Stephens wrote:

So let's just put it out there: Like many RPGs set in a universe with advanced technology, Starfinder has rules for computers, and hacking computers. And, like nearly all RPGs that do so, these rules aren't grounded in reality.

:)

Okay... Let me put it another way... Some would consider the security setup I described as "unrealistically overpowered". Others would call it "realistic, but excessive. Others still would call it "realistic security for a military base". And a small group would call it "unrealistically underpowered".

Whatever the case, just like the quote there by Owen points out; its irrelevant. What matters is being able to design a dungeon that isn't a complete milkrun for a group of ingenious players. While, of course... Isn't so impossible that it doesn't kill them outright.

Also, I actually think the rules as described are actually kind of realistic; albeit simplified, and sped up. Though, even the "sped up" part can have a real world parallel; just that it requires preparation, mostly in the form of a "hacking kit" worth of coding scripts. And last I checked, such a thing is immaterial, and not even worth tracking on a character sheet since it has no physical presence.

The "beat DC by 20 to stop making checks" feels very superfluous and pointless to be honest.

To be even theoretically able to beat DC by 20, you need to be able to pass the DC at a roll of 1.
And, at that point, you autosucceed every check anyway, so you wouldn't likely have to roll in the first place.

I would expect it to be possible to get large a large bonus based on preparation and intell gained ahead of time.

"It is the latest model of Abbadar Corp's Xilene series, with an artificial personality based an old holo-vid star filled with a database of popular culture. We've a complete bio on the owner and I'm working on a personality profile to help anticipate the most likely counter-measures. I'm going to try and hack his receiving records to see if I can't get more information.

Give me a bit more time and I should be ready!"

Something like that could give a bonus where you paid time ahead of time to customize the rig you bring in.

So, could I hack a ship's AI and make the artificial personality really obnoxious?

Mashallah wrote:

The "beat DC by 20 to stop making checks" feels very superfluous and pointless to be honest.

To be even theoretically able to beat DC by 20, you need to be able to pass the DC at a roll of 1.
And, at that point, you autosucceed every check anyway, so you wouldn't likely have to roll in the first place.

I would expect it to be possible to get large a large bonus based on preparation and intell gained ahead of time.

"It is the latest model of Abbadar Corp's Xilene series, with an artificial personality based an old holo-vid star filled with a database of popular culture. We've a complete bio on the owner and I'm working on a personality profile to help anticipate the most likely counter-measures. I'm going to try and hack his receiving records to see if I can't get more information.

Give me a bit more time and I should be ready!"

Something like that could give a bonus where you paid time ahead of time to customize the rig you bring in.

Why am I suddenly picturing things like small single use external drives that exist only to give a boosted performance for only one hour before burning out due to being Overclocked about 1000%?

Actually, something like that probably isn't too different in effect as a potion would be. I get the feeling if we played intrigue games more often, we'd be using potions for a similar purpose on specific skills. Probably to disable traps or to pick locks. Or... More likely given the need to get it right the first time... For fast talking your way out of, or into, a situation.

... Weirdly enough, I somehow imagine that drinking a potion of Charisma right before bluffing is probably going to raise a few eyebrows... But that's neither here nor there.

I wonder what the DC is for a Craft check(s) to build a moon-sized Babbage Engine?

So, could I hack a ship's AI and make the artificial personality really obnoxious?

But maybe the AI can hack your PC's personality and make him/her/they really obnoxious?

Weirdly enough, I somehow imagine that drinking a potion of Charisma right before bluffing is probably going to raise a few eyebrows...

I'm sure it would. Which is why you store the potion in nano-carcerands that are floating in your bloodstream (or stomach or squeedlyspooch or wherever) to be released discreetly upon receiving your mental signal.

What sort of language or script is that text?

Why am I suddenly picturing things like small single use external drives that exist only to give a boosted performance for only one hour before burning out due to being Overclocked about 1000%?

Actually, something like that probably isn't too different in effect as a potion would be.

That's exactly the way I've envisioned it for many years - I *think* that was how it was handled in an old cyberpunk game ... Cyberpunk 2020, maybe? Maybe the Hardwired supplement? At any rate, that's what *I* would like to see and if Starfinder doesn't have something like that in it ... it will once I write it up. Which will be easy to do, just by changing the wording in the Brew Potion feat.

I get the feeling if we played intrigue games more often, we'd be using potions for a similar purpose on specific skills. Probably to disable traps or to pick locks. Or... More likely given the need to get it right the first time... For fast talking your way out of, or into, a situation.

... Weirdly enough, I somehow imagine that drinking a potion of Charisma right before bluffing is probably going to raise a few eyebrows... But that's neither here nor there.

We do that playing Skyrim all the time. :D "Oh, shoot! Forgot to put on Volsung before haggling with the merchant!"

Follow-on ... I think some of this is coming from the old Apple video game Neuromancer. Which was loads of fun. Hacker decks can only hold a certain number of external chips and only run certain levels of programs, based on their complexity/level/tier, so you *MUST* do proper recon to figure out which programs you need to have slotted when you make your run.

(I played loads of different cyberpunk games "back in the day" and liked elements of each system better than others. Odds are very good I'm synthesizing from GURPS, Shadowrun, CP2020 & Hardwired, SWRPG's slicing rules, and at least one other that I can't remember. I know at least one of my players *REALLY REALLY* wants to play a hacker/mechanic, so I've spent loads of time thinking about how to incorporate decent playable hacking into the D20/Pathfinder system.)

But no, the equipment section does not let you buy or sell self-aware creatures, for what I think are fairly obvious reasons.

This is a problem I ran into while designing a playable construct race for Starjammer. Some of the mechanics (healing/repair) are dependent on the construct's cost. So a playable construct has a defined cost. So now I'm wrestling with how *I* want the subject handled in my products before I proceed. New repair mechanic? Self-aware playable constructs are as rare as intelligent magic items and thus have special societal conventions for handling one upon creation? Do they only happen as the result of an Awake Construct spell? Do we introduce the concept of slavery and all that entails? Something else I've not thought of yet?