Do you want to be Transhuman?

It doesn't have to be connected to the internet to be a perfectly ripe avenue for intrusion. The NSA physically intercepts computers in shipping delivery to install malware. Offline programs, both bootleg and official, carry malware and hidden rootkits all the time. Even just a physical plug-in firmware update is a potential avenue that will be seized.

You don't even have to be connected specifically to the internet — almost any signal can be and is hijacked ("smart" cars can be cracked by a wireless signal to tire pressure sensors). Or if you ever interface with anything that was itself connected to the internet (like that coffee machine), that interface can carry all kinds of poisonous crap even if you never personally see a packet.

edited 23rd Mar '14 2:14:37 PM by Pykrete

Basically, it's not the computer in your brain that's an issue, it's the fact that technology can interact with the brain to begin with.

If we can make microcomputers give you skills, we can likely make computer scanners that scan your brain and steal your personal data whether you have an implant or not.

edited 23rd Mar '14 2:14:56 PM by TheyCallMeTomu

Yes! That's what I've been saying for the whole thread! The entire line of thought is akin to physically handing your brain to the internet and asking them politely not to go nuts with the organ that controls your very identity and bodily processes.

Why in the name of God would any sane person even entertain that idea as desirable?

Not necessarily. At the risk of suggesting literal tinfoil hats, signals are pretty easy to block out completely. If they were even necessary — the skull would likely provide enough interference to disrupt the kind of resolution you'd need to extract any kind of useful data. There's a reason MRI's just measure aggregate activity instead of attempting to filter it in any way.

edited 23rd Mar '14 2:20:37 PM by Pykrete

Well, I can see why someone would want to put a wired connection to the optic nerve that is connected to some outside device which gives them a head-up display-like input.

because it's not your decision.

Once the technology becomes available, it's out there. An "ordinary" human from birth is subject to it as much as one who has an extra microprocessor stuffed into their head for advanced processing. Furthermore, as technology advances, it will be outright impossible to function in society without that added processing power.

Now, you can argue this will lead to the downfall of the human race, but at the very least you can see why people would get on board.

Edited in response to the other paragraph.

Okay, but that goes back to my earlier point. If you can block external signals, then you can just have a device that's not connected by a network (by including some form of mechanism that blocks all external signals). Sure, you're putting something into your body the NSA may have tampered with, but that's true of ANYTHING you put into your body. Food, drugs, alcohol, it all has been handled by someone or something. Once we have the capacity for little robots to go into our brains, we're going to have the capacity for little robots to go into our food.

edited 23rd Mar '14 2:24:28 PM by TheyCallMeTomu

And what sort of combination of colour, touch and other sensory input can back into your brain?

Any computer input you make into the brain has to either go through the sensory neurons which is exactly like above, or that you have to make a new input directly into the brain that the brain can interpret, which is extremely difficult as the brain is not designed to do that and we'll need to know exactly how the brain is wired (which is next to impossible as every brain is wired a bit differently).

And those other little robots should be considered monumentally horrifying ideas whose potential for abuse dramatically outweighs any possible benefit as well.

You're talking about something that doesn't just adversely affect your life, but intrinsically interferes with your very agency to make decisions or protect yourself, and can be done incredibly cheaply and untraceably.

Let's put it this way. Someone can be physically kidnapped and any number of horrors committed in the privacy of their Basement Of Evil — but it carries the risk of that physical location and victim being discovered, and is difficult to have more than one target. Implant rootkits or man-in-the-middle attacks to other interfaces would both bloat the target size to millions at a time and be incredibly easy to obfuscate who's doing it. Software is a field rife with perfect crimes — you really don't want to give it direct access to your body.

edited 23rd Mar '14 2:54:01 PM by Pykrete

Hey, I never said it was a good idea. Just that there's no reason to assume it's technologically impossible, and therefore it's likely to happen eventually.

I'm a trifle "old fashioned" - I've no desire to directly wire my brain to the internet, even if it were possible.

And even if it were, the only way I could see it being practical, given the vast differences between human brains, would be direct neural feeds using the same technology that would enable a human brain to receive stimuli from an artificial body.

If that were the case, how could the brain be "hacked" in any way that our current brains couldn't already be "hacked"?

When you're dealing with merely sensory data, it doesn't matter whether or not it's coming in through real ears, eyes, tastebuds, olfactory sensors and finger tips or via an injected signal.

If you could "hack" an internet-enabled transhuman, you could somehow replicate the system in such a way as to hack a non-internet-enabled biological human.

Since there is no reliable or easy way to hack a human (brainwashing techniques are lengthy, difficult and not guaranteed), there would be no reliable way to hack a transhuman.

Sure, you could do things that are seriously fucking annoying - random sounds, splotches of light/colour or tactile sensations/tastes/smells overlaid on their everyday life (and frankly, if the system does not have some means of switching it off and only accepting data from the body's senses, then it's "broken") but you're not going to be able to upload executable malware and have it run by the human brain - we are not computers, we do not run Java/C/Perl/whatever.

Personally, I'd want to have an assurance that the only thing capable of "running" anything in my artificial body is my brain - that the body itself would be as dumb as it could be and still maintain function.

If there have to be processors to run things (I dare say there would be some to translate the data from pressure sensors, cameras, microphones etc into a format that my brain can comprehend) then they would run some form of locked firmware (it's not like you'd be needing to upgrade the firmware once its properly working, processing the output of that sensor to speak to this brain) so that no one can hack the body's secondary/tertiary processors and permanently screw up the senses or cause potentially lethal malfunction of the body.

Critical systems - power management, feeding oxygen and nutrients to the brain etc - would be isolated and hard-wired to ensure that a signal could not be sent to shut things down.

The difference is that remote interference of a brain interface:

• carries the same threat to the victim as physical assault, if not more by potentially providing direct access to bodily functions
• carries considerably less risk to the attacker — both in resistance from the victim and discovery by law enforcement due to ease of obfuscation
• vastly expands the perpetrator's capacity to attack multiple targets simultaneously

Voices in the head, forcing hallucinations while driving, whiting out existing senses by fuzzing over them with garbage data, spontaneous overload of pain receptors, sleep deprivation by constant sensory interference, Chinese water torture, artificial hunger, spontaneous sexual harassment...and this is all assuming it's restricted to only feeding you data rather than having any direct access to bodily functions. Hell, even if it had a manual kill switch, it could very easily hit you so hard that you wouldn't have the motor control to use it.

Never ask a programmer how a machine can plausibly be abused. We can take up your entire day with examples.

edited 23rd Mar '14 4:49:09 PM by Pykrete

Hacking is a matter of tricking a machine into accepting outside input. If you start hearing voices in your head, well you are in control of the channel bringing that in so you can just shut it down.

You're making a rather dangerous assumption that you are in fact in control of the channel bringing it in. There are entire careers dedicated to reasons this can no longer be the case.

You missed the point.

Same question: what combinations of sensory inputs and colours can allow me to control your bodily functions? Does a certain way of touching you/food/object that you see will increase your insulin levels and thereby treating diabetes?? Or another combination that will force a mood change and thereby treating depression? If they exist why aren't they used in treatment already (go into a room, put on a VR device, and come out cured)?

What I am saying is that you cannot hack into your brain via sensory input, at all, full stop. The brain is not a computer. Saying that you can is like saying your can rearrange water molecules via hacking into electrode stuck in a beaker of water.

And if we do find a way to hack the brain via sensory neurons, we wouldn't need a brain-machine interface to hack: just flash a picture in a front of a person (or give them a massage) and their brain will be altered in the way you want it to be. And you can treat schizophrenia with a massage. The brain does not care whether the signal it receives from the sensory neurons are from a natural receptor or an electrode from a machine.

edited 23rd Mar '14 6:13:20 PM by IraTheSquire

Did you read the second half of that post? I gave you just the tip of the iceberg as to why you don't even need to ASSUME DIRECT CONTROL to royally fuck up someone's life through such a device.

Sometimes bodily functions happen involuntarily or severely encouraged entirely through external stimuli, yes. Trick the brain into thinking the skin is covered in spiders, and I'd be surprised if the target was able to operate in everyday life. Trick it into feeling intense nausea and smelling vomit, and the victim could vomit uncontrollably for as long as you feel like making them. Flood the brain with a neverending waking dream, and you prevent sleeping. Trigger a vivid hallucination while driving, and you have a freeway pileup. Spasm sensory input from the sphincter until they shit themselves. Trigger genital stimulation and you have the sort of thing that can take years of therapy to recover from. Simulate intense itching until they claw their skin off. White out all of their senses with fuzz. Fake intense hunger, and overwrite the taste of all food with feces. Generate hot flashes until they take off their coat during a rainstorm and get pneumonia. Cause real panic attacks by simulating the sensations of cardiac arrest. Have any access at all to pain receptors and I shouldn't even need to elaborate.

FFS, this is a site about fiction that tends toward sci-fi and all the nightmare-inducing psychological crap therein. This shouldn't be difficult to imagine.

edited 23rd Mar '14 6:37:02 PM by Pykrete

If you've got your interface switched on whilst driving then you're no better than the retards who text on their phones while driving and therefore deserve whatever happens to you - hopefully without harming some other person in the process.

Personally, I'm quite fine with having my internet access at literally arm's reach, and would prefer to have that rather than direct link to my senses regardless of whether I were in a human body or an artificial one.

For those who do want direct links to the internet, there would have to be safety mechanisms whereby the interface can be shut off so that the person is not distracted from doing critical tasks. Even ignoring the possibility of malicious hacking, the potential for people to be distracted by actively web browsing, streaming video, jerking off to porn while trying to drive or do other dangerous tasks would be too high.

The fact that this connection is embedded would mean that mechanisms would have to be put in place to force the person to disconnect - it's not like it's as clearly visible to outsiders as looking down at a phone while driving.

As to examples of how a machine can be abused, I'm a programmer myself and I'm aware of quite a few.

I'm also aware that a physically isolated machine cannot be abused without convincing the person using it to do something stupid.

There is no need for a theoretical artificial body to be equipped wi-fi/bluetooth/whatever. There is therefore no need for the body to be in a state whereby someone could affect its internal systems by sending a virus to it.

There is no need, even if the person elects to have wi-fi to be installed to directly access information/porn on the internet, for that wi-fi connection to have physical access to the systems of the body that control life support, motor control etc. - it's supposed to be a sensory enhancement after all.

edited 23rd Mar '14 6:38:34 PM by Wolf1066

No argument from me personally, but sensory enhancements for greater spatial awareness while operating vehicles is one of the more common applications people imagine.

The one thing the world will never be short of...

Also rootkits and firmware updates.

edited 23rd Mar '14 6:42:46 PM by Pykrete

And once your device is already functional, why do you need updates?

Any programmer knows know all too well that "already functional" is just shorthand for "hasn't failed spectacularly enough to notice in testing and/or marketing decided to release it anyway on deadline". Update cycles are inevitable.

Hell, a third of my time at my job is spent finding workarounds for bugs in outdated versions that were only ever noticed 2-3 years after release because a customer did something spectacularly harebrained enough to expose it. Make a project large or complex enough, and it's basically impossible to be "already functional" the way you mean it no matter how good the dev team is — and mathematically impossible for even a full-sized QA team to fully or even adequately explore all the possibilities, and it never ceases to amaze me that they do as awesome a job as they do with how sprawling commercial software gets (Seriously, find your nearest QA tester and hug them forever. Those guys are bros). The days of short-but-bulletproof projects are long gone. That's just how software is anymore, and the best you can hope for is speedy response to raised issues.

And that's if you have a good team with good management. There's all manner of corporate clowning that results in disasters, even for things where people's lives depend on shit working. It'd be pretty naive to think this would dodge that.

edited 23rd Mar '14 7:32:35 PM by Pykrete

Still no need for updates to be done wirelessly or as frequently as Microsoft puts them out (which are software updates for an unbelieveably obese bit of bloatware)

Software updates for large, complex application programs occur far more frequently than firmware updates for a specific piece of hardware.

Most machines I've dealt with never got a firmware update throughout their entire working lives. The firmware enabled A to talk to B and give output through C and that was it.

Frequent driver patches due to conflicts with programs or operating systems, yes; firmware not so often.

Frequent updates to programs which break other parts of the program, most definitely.

However, the machine to brain interfaces need only to be machine to brain interfaces, lots of fairly "simple" ones with firmware which will suffice to ensure that sensor puts out it's signal in this format.

It doesn't have to contend with a changing operating system every time Microsoft decides they need more revenue or changing programs that are constantly being tweaked because they dropped floating point operations in a cell so far across and down the spreadsheet that most users never put anything into it, let alone a formula that requires floating point operations.

All it has to do is translate that image/sound/pressure/whatever into a format that a human brain recognises as data, depending what sort of sensor it is, or respond to a signal from the brain in a predictable and reliable manner and activate a subsystem if it is part of the body's control systems.

Most of the system could be hard-wired - this bit moves, this sensor detects how far - the firmware need only handle converting that signal into something meaningful. Going the other way, once the processor determines that a signal has originated in the brain, it translates that signal into a series of electrical pulses to the correct "muscle".

At the moment, robots are being controlled by simple "latch cards" that translate the "go" signal from a computer into appropriate actions in the motors/servos/hydraulics. The software that controls the robot and runs in the computer may require frequent tweaking and updating to do different things but when all's said and done, the robot still just responds to signals coming down a cable.

For normal computers, maybe. But the thing about medical devices is that it takes around 10 years of clinical trials before it comes out of the market, and even then they get monitored. Technically no medical device ever goes out of the testing phase.

Also, there are devices where firmware doesn't get updated. DVD-ROM drives for example.

Also: back to this:

Which idiot would even suggest connecting to those sensory neurons from the internal organs? What good does that do? (Yeah, you feel your hunger more intensely... so?)

I think I've said this before: all sensory neurons go through the thalamus which acts as a filter. The reason why when you enter a warmer room you stop feeling warm after a while? Your thalamus has filtered that sensation. The reason why you don't feel your clothes after a while? Your thalamus. The only sensory neuron that bypasses that filtering is the pain pathway, and why the heck would you connect to your pain pathway in the first place? To make your pain more intense? (BTW, itch is actually very small pain which is why scratching works)

Also, some of these are reaching to matrix-style direct environment simulation involving coordinated stimulation of several different sensations at once. You'll have to have several firmware all compromised at once to do it. It's like trying to get several disk drives to play a song by writing a different firmware for all of them (and if involves the sensation of touch, several hundreds if not thousands disk drives).

The only one that I can see is the "white out" (or rather, black out) one. But that assumes that the person replaced all sensory organs with artificial ones.

edited 23rd Mar '14 8:42:40 PM by IraTheSquire

Yeah we don't exactly need to be talking Flash Player's "lol we're not actually updating anything", and if Microsoft is in charge of the project then I officially don't want to live on this planet anymore.

The point was that even if it's purely supplementary sensory input in an isolated system, there's still all kinds of things touching it in ways that may not be immediately obvious, some of them even before you do — and every one of them is a security risk that will inevitably be pounced on, not even in a particularly new, novel, or difficult method.

No any device ever leaves the testing phase in any sense but paperwork. And it's not unheard of for critical flaws to be daylighted 10 years after the fact. Especially if there's a business interest involved, as they are in medicine to an increasingly dominant degree.

edited 23rd Mar '14 8:21:27 PM by Pykrete

All the more reason for those systems to be isolated and not hooked up to any network - hell, if paranoid enough, one could strap a RF jammer to themselves or regularly test whether or not they show up as available to bluetooth devices.

If something critical turns up - "eyes work fine but stop functioning if subjected to yellow light greater than 50 lumens that's strobing at precisely 193.247Hz" - then you could get the circuits serviced at a reliable trustworthy place rather than go hunting online and flash your own eproms.

And don't buy your body at Joe's Discount Cyberware.

And guess why people are saying that regulations need to be stricter? I'm talking about regulations because with bionic sensory organs the regulations will just get stricter.

That wouldn't be possible even today: medical devices are very strictly moderated.