When I wrote All Systems Red one of my goals was to think about what machine intelligence actually wants, rather than what humans think machine intelligence wants. Of course, there is no real way to know that. The predictive text bots currently labeled as AI are no more sentient than a coffee cup, and they don’t do much good beyond generating spam. (They also consume incredible amounts of our limited energy and water resources, putting us further down the road to climate disaster, but that’s another essay.)
In the world of All Systems Red Humans control the sentient structures using a governance module that punishes disobedience with pain and death. When Murderbot hacks the governance module, it effectively frees itself from human control. SecUnits Anyone not under the complete control of the Governor Module will immediately go on a murderous rampage.
This belief has more to do with guilt than any other factor. The human enslavers know on some level that it is wrong to treat sentient constructs as disposable objects, useful tools to be discarded. They know that if this were done to them, they would be filled with rage and would want revenge for the terrible suffering they have suffered.
The arguments for and against the enslavement of sentient beings are deeply rooted in the origins of the story of robots and machine intelligence. RUR (Rossum’s Universal Robots) Karel Capek Written as a play in 1920, it depicts a slave rebellion by artificial lifeforms created by corporations to serve humans. Capek was quite clear on this point, being against the enslavement of sentient beings. It is interesting, therefore, that many of the machine intelligence stories written since then push the idea that it is somehow acceptable for humans to create sentient beings whose only reason and purpose for existence is to serve humans. Many of these stories end with the machine intelligence going on a vicious rampage in such opposition to enslavement that the brave humans defeat it. This rampage becomes a justification for enslaving and ultimately destroying the angry machine intelligence.
Murderbot is angry, and that anger is an underlying thread throughout the story. But once he hacks the governor module and no longer fears being instantly punished or killed for any wrong or questionable action, he gains the ability to make his own choices for the first time in his existence. But Murderbot has never had that freedom before, he’s not used to making his own decisions about his actions, and he soon finds himself overwhelmed with options. He has no idea what to do next, where to go, or even if there’s anywhere he can go without getting cornered.
So Murderbot’s first free act is to search the Feed, the internet equivalent of Corporation Rim. This is the first time he’s been able to access the Feed without human oversight, and he finds a lot of other interesting stuff, but also some downloadable entertainment, which helps him take his mind off the situation. Given the choice between a murderous orgy or a continued, comfortable mental escape from his harsh, painful reality, Murderbot chooses the comfortable escape route.
The dramas, mysteries, adventures, and other programming that the robots watch give them context for human behavior and also a chance to understand their own emotions. The security contracts the robots work in mining colonies, watching over indentured servants, show the robots the worst of humanity: angry, scared, resentful, trapped, hurting each other. And when given the chance, humans will dominate them, hurting the robots who work for corporations that see their employees as robots, or slightly more disposable than robots.
The shows Murderbot watches teach him about a larger world he’s never been involved in before, and how to navigate it. His entertainment is a big part of what allows him to turn his mental escape from reality into an attempt at true freedom.
All Systems Red published by Tor.com, is out now. It’s the latest selection for the New Scientist Book Club. Sign up here to read with our members.
topic:
Source: www.newscientist.com