A Request of My “Cleansed By Fire” Readers

novel_handlightAs Big Man noted in the comments section for part 36 of the novel, some aspects of the artificial intelligence systems in the novel have been confusing, both from the aspect of their reproductive capabilities and the differences between an primary AI, secondary AI and DI.

I liked his idea of adding in some throaway scenes in the first few chapters of the novel to slowly fill in some of this knowledge so that some of the later stuff would make more sense.

Looking toward the eventual rewrite of this novel when I get to the end of it in this online/blog form, I have crafted four rough scenes. I would like to submit them for comment and suggestions here. And as much as I love Big Man’s input, because he’s helped a lot with some things over the months as I write this, I would love to hear from the rest of you, too. 😉

(For those of you who don’t give a flying bout of fornication about the novel, I will likely have some kind of other post up this afternoon or evening.)


One of the minor deacons, Frederich, caught Gregory’s attention from a side hallway, and sprinted over before the Peteris could be whisked away to a meeting or to deal with some church crisis.

“Peteris, I just wanted to let you know that Ghost’s first DI looks to be ready to expire officially soon. Probably within hours.”

Gregory frowned. Ghost, the UFC’s most powerful artificial intelligence, and its only primary-class one, didn’t talk about her two offspring much, and he wasn’t sure how the news might affect her mood today—or if it would at all. All things considered, Biblios had lived a long life for a demi-intelligence; almost as long as the average secondary-class AI—which was just a little less than a human with an unhealthy lifestyle and no interest in obtaining routine preventive medical care.

A shame though; the younger children really enjoyed how Biblios told them stories in both their religious and secular classes, Gregory mused. As much as it would have been nice had Ghost produced a secondary AI with a strong theological or educational template, her first-born DI had served well.

His systems had begun to degrade two years ago though, and he’d been off active duty for the past four months as his self-repair systems continued to fragment at geometric rate. No one had ever really settled on an official doctrinal answer as to whether AIs and DIs had souls, but the UFC had always erred on the side of caution, and Gregory intended to do the same.

“Thank you, Frederich. Could you work with Annah to write up a short eulogy and memorium prayer so that we can honor Biblios properly at the next public service?”

“Indeed, Peteris. And thank you for including me.”

Gregory put his hand on Frederich’s shoulder and gave a quick squeeze. “Don’t thank me too soon. You’re our next-best storyteller after Biblios, so I’m probably going to make you handle tale-time with the youngsters for a while.”

Frederich smiled. “I suspected as much. Don’t worry, I bought a ready supply of anti-anxiety tabs and headache suppressors in anticipation of that.”


Daniel Coxe was still trying to figure out what to do with his newfound knowledge of the Godhead’s clandestine actions, and was not happy to be distracted with co-workers and business matters when he needed to focus on how to get out of the Catholic Union without being noticed.

But as with so many things in life, he noted to himself, desire and reality don’t often converge.

And in this case, reality was Oliveri Marschone, a virtual defense tech four levels below Daniel’s own station, reminding him of what he saw as a need to upgrade the Godhead’s electronic and virtual security. Fear and confusion made Daniel even less polite than he might have been otherwise—and rarely did Oliveri inspire Daniel to politeness.

“Ollie, the only reason you’re bringing this up, for the third time in two weeks and just a week after our last upgrade, is because you want to feel important, and you think that haranguing me is going to get your level bumped up,” Daniel snapped. “It won’t, and if you bring it up one more time, I will do my level best to push you down two more grades.”

“Daniel, the Godhead is the most…”

“…important AI in the Catholic Union. The crown jewel of the Vatican,” Daniel finished for him. “I’m in the Godhead’s core systems regularly. It’s my sole job to keep him healthy and safe. A nanomite couldn’t penetrate the system protections we have in place if someone dropped one right inside the core processors. The firewalls, virtual turrets and warware we have in place could fight off hacks, buzzbugs, viruses and anything else thrown at us if every nation on Earth decided to pelt the Godhead with hostile intent over the SystemGrid simultaneously.”

“You can never be…”

“…too careful, Ollie? I am too careful to a fault. Bugger off and harass someone from physical security. We have this huge complex to protect the Godhead from virtual and physical attack, and I’ve seen at least two sliptrucks come in without being inspected since October, and there was some confused courier that someone had directed straight to my office back in June. At this point, I’ve worried about bombs and matter-eaters getting into the complex.”

Daniel turned and stormed off without another word.

It might almost be worth risking another day or two here in the Union just to write that wanker up, he thought, if I wasn’t so utterly fond of my own skin, that is


“Ather, I hear that you’re getting a virtual retard to monitor your inquisition tools,” taunted Yuri man-Juris as Ather approached the Pit in the command center, where the ranking communications and logistics officers kept the templars organized. Yuri was, by all accounts, one of the next in line for promotion to sup-Juris, though Ather couldn’t imagine why—aside from the fact his mother was a corporate top-hat in the armaments market.

“Yes, Yuri, would you like to serve as a surrogate prisoner for the test runs? Since it will only be a DI instead of a secondary-class AI now running the systems, she probably won’t be able to get past level 6 pain thresholds during an interrogation.”

With that, Yuri suddenly remembered an important appointment with one of the comm-log techs, and Ather grinned to himself. Honestly, he had his own reservations about using a DI instead of an AI to manage the inquisition hardware for the more recalcitrant suspects sent Ather’s way, but there was a clear cost advantage.

Besides, AIs have that annoying tendency to make suggestions about how I could do my job better, even though they don’t know the first thing about feeling physical pain, he considered. DIs don’t have enough cognitive power for strong opinions. Better yet, they have shorter lifespans, so all the sooner that we can discard them and all those unsavory encounters they have committed to their memories


Lyseena frowned at the requisition Willem Staffordis had set before her.

“They want to have Guardian attempt to sire a primary AI with Ranger?” she asked. “Did anyone say why? None of the primary AIs for the regional templar offices in any part of the Catholic Union are older than three centuries. Short of being taken out with a missile barrage, they should be solid for at least the next 2,000 years. Is the military looking to improve the caliber of their AIs by riding on our technical successes?”

“Possibly,” Willem said. “But I got the impression from the Black Pope’s liaison they simply want to see how Guardian’s tactical template meshes with Ranger’s investigative template. No one’s ever managed successful procreation with those two templates before, but there’s also no indication that the two are incompatible. I think someone is just hoping that they can create a novel hybrid template and look important. Worst case scenario, someone will end up with a solid tactical or investigative AI. Ranger and Guardian both have perfect track records with their ability to spawn both primary and secondary AIs. Not a single DI for either of them.”

“Somebody wants a primary AI with something midway between tactical and investigative, they should do some heavy lifting and design it from scratch. But approve it, then, Willem,” she said. “It’s no skin off my hindquarters either way, and I have far too much to deal with getting ready to oversee the Grand Requiem and the millennial events to worry about whether a couple of our AIs are going to have a virtual romp to make a bouncing baby AI.”


6 Responses to “A Request of My “Cleansed By Fire” Readers”

  1. February 6, 2009 at 12:12 pm

    I like those scenes.

    Why do DIs only live as long as humans? I think you said elsewhere that AIs can live thousands of years, so I would assume that a DI could live a few hundred, even if it is a damaged reproduction.

    Also, is it impossible for humans to repair AIs and DIs? I know you said the AIs are self-repairing, but I was wondering the humans didn’t help the DI repair himself if they wanted it to survive longer.

  2. 2 Deacon Blue
    February 6, 2009 at 1:49 pm

    Primary AIs can live thousands of years potentially because they are very carefully crafted and are given lots of room to store their memories and personality databases. And when they spawn another primary AI, the two parents make very specific inception routines to help ensure that their spawn is comparable in quality to themselves, and has a similar kind of complexity and self-repair capability.

    Secondary AIs, while they also are powerful and have distinct personalities, cannot be afforded the same kind of dedication to very complex systems and larger databases, for a number of reasons:

    1. Cost.

    2. Practicality. Secondary AIs are simpler, and designed for simpler tasks, so a 50-120 year life span is sufficient and, in some cases, can actually be TOO long depending on the AI’s function. Most secondary AIs have tasks like office management, acting, teaching, musical performing, virtual sex work, accounting, medical assisting, etc. A primary AI is a computational and multitasking beast in comparison to a secondary AI. For example, Ghost runs much of the functions of the UFC church across an entire solar system, while handling theological ponderings, managing theological records, and keeping an eye out for intelligence related to the Vatican and its attempts to destroy the UFC.

    If you think of a secondary AI vs. a primary AI vs. a human, look at it this way. A secondary AI is like having a human worker who doesn’t need to sleep and is faster. Humans are still necessary because, well, secondary AIs still lack a certain human spark and their creativity is lesser than that of a human. Humans are also more cost-effective for a lot of jobs compared to having a bunch of AIs to do it. A primary AI is superhuman, able to do the tasks of an entire team of humans (but still, lacking in social graces at times, depending on how well the AI has developed his or her personality, and still not quite as creative as a human in some cases….and WAY more costly than a team of humans.)

    3. Given that they don’t have systems that are generally as spacious (data-wise) or as complex (for multiple redundancies) as primary AIs, the older they get, the more clogged up secondary AIs get and the more unstable at a certain critical point as their data begins to degrade. The fact they have personalities at all (even if they aren’t as deep as a primary AI’s or a human’s) begins to become a handicap. Think of how long you can get out of a PC before it becomes obsolete. Secondary AIs aren’t anywhere near THAT limited, but they just aren’t designed for human-level lifespans. Even in 4001, humans don’t know enough about how personality is generated and memories are efficiently stored to do it well enough for a secondary AI to have a longer lifespan. The virtual personality databases are unwieldy data hogs compared to the efficient, organic human brain.

    DIs, because they are essentially cognitively impaired compared to other AIs (whether primary or secondary) have their problems compounded and their lifespan reduced accordingly. But they are still desirable for many functions because they are less costly to maintain than even a secondary AI is. DIs are like a generic/budget AI.

    As for humans going in and fixing AIs that begin to reach the end of their lifespan, it’s sort of like us today trying to go in and fix someone with Alzheimer’s disease. We just don’t have any way to do it. Humans can fix certain types of damage to AIs or DIs, but not all types. In creating computers with human-like characteristics, there eventually come to be problems that cannot be cured at some point in an AI’s lifetime.

    With primary AIs, this problem hasn’t yet come up, so the maximum lifespan of a primary AI isn’t known. But they are expensive to maintain, so people don’t need or want as many of them anyway, compared to secondary AIs or even DIs (which have less ability to multitask or take on multiple roles, but are still better than a mere computer).

  3. February 6, 2009 at 5:06 pm

    YOu should get that Alzheimer’s explanation into the story somewhere. something that lets the readers know why repairs are not feasible because for the average person when a machine breaks we expect it to get fixed.

  4. 4 Deacon Blue
    February 7, 2009 at 4:58 pm

    That’s a really good suggestion, Big Man. I’ve added it into into one of the scenes above as of last night thanks to you, but since wireless is still down (and I’m on the wife’s PC), I’ll have to wait until I can get my own computer to share it here and see what you think.

    And hey, no comments from anyone else. Is Big Man my biggest fan? I hope he’s not my only one.

  5. 5 Deacon Blue
    February 8, 2009 at 12:36 am

    So, in the first “throwaway scene” I figured I could change the fifth paragraph like this and get the Alzheimer’s example in there:

    His systems had begun to degrade two years ago though, and he’d been off active duty for the past four months as his self-repair systems continued to fragment at geometric rate. A process as relentless, incurable and fatal as various dementias like Alzheimer’s disease had once been for humans–and inevitable for all DIs and secondary AIs. Whether primary AIs were immune or not only time would tell, but the oldest primaries were now well over a thousand years old and still going strong.

    Again, thanks for the suggestion Big Man. It was a good thing to point out. I hadn’t been thinking of AIs as simple machines, but many readers would.

    I guess the closest example I can think of in modern sci-fi would be the replicants in the film Blade Runner. They were manufactured humans, but they had a built in expiration date and no amount of tinkering with them was able to stop or reverse that expiration date.

    In the case of my AIs, it’s not a set expiration but still a limited lifespan compared to humans. Because my AIs in this novel are, while not human, still not simply machines. They are mankind’s attempt to create life from computers, in addition to doing so genetically with neo sapiens.

  6. February 24, 2009 at 3:43 am

    Hmm, very cognitive post.
    Is this theme good unough for the Digg?

Comments are currently closed.

Deacon Blue is the blogging persona of editor and writer Jeffrey Bouley. The opinions of Jeff himself on this blog, and those expressed as Deacon Blue, in NO WAY should be construed as the opinions of anyone with whom he has worked, currently works, or will work with in the future. They are personal opinions and views, and are sometimes, frankly, expressed in more outrageous terms than I truly feel most days.

Jeff Bouley


Jeff Bouley

To find out more about me professionally, click here. To find out more about me generally, click here.



You can reach Deacon Blue/Jeff Bouley at deaconbluemail@gmail.com.



For my public profile, click here.


Tales of the Whethermen

My superhero fiction blog, click here


Raising the Goddess

My parenting blog, click here

Copyright Info and Images

For more about images used on this site, and copyrights regarding them, as well as usage/copyright information about my own writing as posted here, click here.

Deac Tweets


Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 833 other subscribers
February 2009

%d bloggers like this: