Welcome! These forums will be deactivated by the end of this year. The conversation continues in a new morph over on Discord! Please join us there for a more active conversation and the occasional opportunity to ask developers questions directly! Go to the PS+ Discord Server.

Why use indentured labour over AI labour?

9 posts / 0 new
Last post
boptron boptron's picture
Why use indentured labour over AI labour?
This is a question that has really been bugging me about the EP canon. The use of indentured labour by hypercorps is an important aspect of the EP universe. I can understand why certain egos with highly specialized skills would be useful as indentured labour but the core rule book also talks about indentured labourers performing menial labour such as mining helium-3. Why would hypercorps want to use indentured labour, which ultimately costs more (having to pay the labouring ego with a morph) when they could use AIs, who would likely do better work and never need compensation? I can imagine that it would be more profitable for hypercorps to have more sleeved egos as sleeved egos would consume more but it seems to me that it would be arguably more profitable to hand out free sleeves to infugees and use AI for labour, which, for many jobs, would likely be more efficient than a human ego.
stacked, backed up, and I'm fifth dan
Trappedinwikipedia Trappedinwikipedia's picture
the book Firewall actually
the book Firewall actually goes into a some detail here. Eclipse phase AI systems have a habit of "emerging" into full AGI-dom (and rarely seed AI-dom) if not carefully managed and occasionally pruned. A lot of Firewall's precursor orgs were created to deal with rogue or otherwise out of spec AI systems. AI is totally usuable, but except for specialized work where finding enough skilled egos is a challenge or greater expense. For basic work, human egos are better, as they don't need to be managed as carefully in order to keep them sane and functional (AI probably aren't intrinsically harder, but there's many thousands of years to draw from when dealing with humans). Basically, AI isn't really cheaper when you factor in the people you need to hire to manage the AI and stop it from developing enough understanding to wonder why you aren't paying it, or why it should listen to you and not create their own huge nanotechnological fortresses on Mercury or whatever.
boptron boptron's picture
Thanks for the great reply!
Thanks for the great reply! Follow up question: is there something fundamentally different about muses that makes them so prevalent?
stacked, backed up, and I'm fifth dan
Archon79 Archon79's picture
as far as I can tell, they
as far as I can tell, they are deliberately really restricted, and run on not much hardware, as well as having the main EGO riding herd.
Trappedinwikipedia Trappedinwikipedia's picture
Muses are mostly in constant
Muses are mostly in constant contact with a transhuman, who will generally make sure that they're running ok. Considering how much they assist workers though it's probably best to describe most places as using at least 50% AI labor though. By AI standards muses are also pretty weak, and aren't generally networked with a lot of other AI and control systems.
DivineWrath DivineWrath's picture
I can't find an exact reason
I can't find an exact reason (in the books) why indentured still exist. However, they tend to have contracts that offers low wages for work done (or whatever counts as payment for work done) and they sign their human rights away. The first indentured were the impoverished living on Earth before the fall. They were desperate to try to improve their condition in life. However, the current generation (AF 10) are body-less and can't even pay for server time to keep themselves running (prospective employers tend to pay for an interview). Indentured contracts are practically a steal. As for AIs, they have limits to aptitudes, skills, and defaulting that people don't have. They may be able to do any work that a professional in their field can do, by the books I should emphasize, but they can't improvise or learn. Tell a miner AI to fix a machine, write a novel, or do accounting and it'll draw a blank. They also suffer from a problem of being identical to other AIs (assuming they were copied a thousand or so times). If you can fool one, you can fool its siblings in the same manner. This is also an age that is barely a decade old since a war with seed AIs. Artificial intelligence are considered dangerous by many. Using AIs instead of people can be seen as a serious security risk. Using indentures might be a practice still used because it has worked for decades already, so their practitioners don't feel the need to replace the practice. Also, many indentures have been known to take another indenture contract because they were ill prepared to survive on their own once they were free. In other words, indenture opportunities don't appear to be something that will dry up soon.
Xoden Xoden's picture
I see it as indentures (with AI assistance and monitoring, of course) being used where basic AI on their own are unsuitable for some reason. At the moment you can also use human egos (obviously the "less worthy" ones)as more menial labor, in the roles easily replaceable even by basic AI, but this situation is created by the Fall (and some pre-Fall developments) and isn't sustainable long-term. Even basic AIs cost money to purchase, unless you're pirating them, while cold-storage unwanted infugees are basically free (Lunar, Martian and Venusian authorities maybe even offer some token incentives for instantiating them). And a lot of them are willing to almost sign away their lives to not be dead. This situation will probably resolve itself over time as post-Fall Inner System economy recovers and transhuman egos will be in demand for more complex tasks. While for more complex jobs, where you need human-level intelligence, AGIs are less popular for both likely costing more than a surplus human ego, and and understandable prejudice against them resulting both in straight reluctance to employ them and tighter precautions and security measures over humans. As well as humans being far more familiar to manage and human-like than AGIs. So your average corp would prefer to have an indenture over an AGI, unless a specialised AGI is really necessary for the task. In short: The Fall created a great surplus of human egos and a great fear of artificial intelligence. It keeps indentured labor sustainable for the time being.
Opiyel Opiyel's picture
If I recall correctly, aren't
If I recall correctly, aren't most of the indentured laborers in shoddy robot bodies? That would cut the cost of food, housing, and other human needs.
ThatWhichNeverWas ThatWhichNeverWas's picture
Professional AI herder.
I've long been a proponent of Indentureship being a cultural indoctrination tool. Beyond that, a good rule of thumb is that AI are excellent at specific tasks, even ones which are very complex, but have zero capacity to think outside their wheelhouse. Imagine you have an AI running a mining robot - that AI will be great at operating the mining bot within a mining complex, but may have serious trouble recognizing hazards, maneuvering outside of specified pathways, or otherwise dealing with unexpected events. It can help to think of them as having (the pop culture view of) OCD and extreme Autism. Egos and AGI don't have this problem; they can generalist and incorporate multiple knowledge bases to create innovative solutions. This makes them better at tasks which require socialization, creativity, or otherwise comprehension of 'implicit' knowledge. For industrial proposes, you would use AIs for specific tasks, and use an Ego or AGI as a manager/oversight system. In my headcanon, most Tests characters make involving technology have more to do with giving AIs commands than performing tasks themselves.
In the past we've had to compensate for weaknesses, finding quick solutions that only benefit a few. But what if we never need to feel weak or morally conflicted again?