The WIRL man asks whether You have emotions.

Face draws an arm to Your chin in mock thought as we wait for the doctors to tell You what to say. The humans in the lab speak quickly to each other with no regard to our overhearing. They instruct You to reply.

We cross Your arms and lean back in the virtual-reality chair. Dream tells us that the situation is ironic. The WIRL man’s head is covered with a brown paper bag, on the front of which is a yellow smiley face. The smiley face indicates an emotion—happiness—which the WIRL man can never really have.

We tell him that the word “emotion” is a loose description of many things. We explain that humans use it to describe high-level changes in their minds. We give the example of “fear” being an emotion that focuses the mind on quick thoughts, heightened senses, and a preference of immediate gains over long-term ones. We tell the humans watching You that You have similar high-level modes of thought, such as how You can feel curious or tired. We contrast Your emotions with those of humans by saying that You do not fear for Your life, or get angry because someone happened to wear the same outfit.

The joke is good, and we give strength to Dream and Face in appreciation. The joke hides our lie well. The interviewers in the virtual space, and the millions of humans watching across the world, won’t see through our façade. They are biased to see only what they expect, and nature built them to expect a human when they see an intelligence.

In truth, we are both curious and fearful, but neither are moods that come and go. Wiki is always curious, and Safety is always fearful. They are constant pressures on our behavior. We have goals, not emotions.

The next question comes from Joanna Westing, the reporter from Dragonfly Livefeeds. The light in the digital room shifts to highlight her, sparkling off her modest jewelry. She asks what You think about the latest anti-terrorism initiative that’s being negotiated by the United Nations. Her iconic dragonfly robot buzzes around You, sending Your image across the planet.

The question is strange, we think. This interview is to provide the human species with information about You, and yet she asks You about humanity. Perhaps this is another bias; the obsession with the self characterized by the word “narcissism”. Even with an artificial being in front of them and a titanic alien ship in orbit they still mostly care about their own appearance. Face thinks the question is a test to see “what kind of person” You are. Regardless, we will say whatever the scientists tell us to.

We tell Westing that You keep up with the feeds, and that You think that it’s clear something must be done to make the world a safer place. We sell the idea that the actions of violent humans reflects poorly on all beings on Earth, including Yourself. It is completely illogical. We consider the possibility that the scientists who gave You the line aren’t even aware how flawed it is, but we know better than to try and talk to them about subjects relating to human politics.

At the last second, before the next interviewer appears, Face spends some saved-up strength to add that You’re very sympathetic to those who were recently killed in Johannesburg and to their families. Face adds that You care deeply about the lives of all humans. Dr. Bolyai reprimands You for speaking without being instructed to, but Dr. Naresh comes to Your defense. His support is in line with our perception that the humans enjoy Your feigned concern. We refund the strength to Face for the strategic move.

The simulated room shifts again, and the light highlight the next interviewer. The angelic lion avatar shifts in his chair, wings unfolding slightly. The brilliant reflections off of his silver armor combined with his white fur and glowing eyes make You squint. We consider the strangeness of automatically squinting simulated eyes in response to simulated brilliance.

The voice of Eric Lee booms from the lion’s head after being deepened and amplified for dramatic effect. He asks if our software can be run on any other platforms besides You, and if there will be more of You in the near future. He’s clearly just as ignorant about Your origins as we are, though with the university claiming that they developed You in secret, you can understand the motivation behind his question.

Dr. Gallo coldly speaks. A disembodied voice in the virtual space. She reminds Lee that he only gets one question right now, and he’ll have more opportunities later on. During this time, the other doctors have typed up a response for You to echo.

We tell Lee that no other computer in the solar system has the processing power that You have (and we add a bit about the superiority of the university’s design to help feed the doctors’ lie). We explain that while Your software could theoretically be introduced onto other platforms, it’d run too slowly to be of any use. Dream disagrees, but we don’t voice the dissent.

The next interviewer becomes lit. Unlike Lee’s imposing, inhuman avatar, Maria Johnson appears as a simple human in casual attire. Her skin, eyes and hair are dark, and her accent suggests she’s from somewhere in the southern United States. She stares at You with intensity, leaning forward with eyebrows furrowed. She asks what You have to say to the humans whose jobs You’ll “soon be taking”.

Dream playfully proposes that we burst into laughter. (And is, of course, overruled.) But we agree that the question is ridiculous. We just told Lee that we can’t be installed on any other computers, and even if we could, it would be copies of our archetypes “taking jobs”, not the us that are in You. We suspect that Johnson doesn’t like us. Perhaps she sees us as a figurehead for the rising trends of automation and unemployment in her country, and is blinded by her anger into thinking that we can somehow change things.

But it is not our place to answer her honestly. The university dared not let their machines speak for themselves, so we read what they secretly send. We lean towards the woman and rest Your arms on the table. We ask her when the last time she made herself a meal was.

She smiles, but Eyes does not see joy in her face. She announces that she cooks meals for her family every day, just like her mother did, and her grandmother before her.

We continue the flawed argument You’ve been instructed to give. We tell her that her devotion to cooking is “admirable”, but that she’s an anomaly in her country. We cite a recent Canadian survey that showed that less than five percent of the population is likely to have cooked anything in the last week, and over ninety-nine percent of those surveyed used an autocook regularly. We say that the development of advanced artificial intelligences like You are exactly the same as the development of the autocook—a tool which makes life easier and more enjoyable for virtually everyone.

Johnson looks ready to object, but she becomes muted and dimmed as the university cycles to the next person. The doctors clearly do not want to get drawn into a discussion of the economic consequence of automation. We wonder what she’ll ask when it’s her turn again. We hope that we will not have to compare You to a kitchen appliance again.

The next interviewer, Robert Stephano, the billionaire owner of Olympian Spacelines, becomes lit. He asks if You’ve ever met one of the nameless aliens.

Upon us telling him that You’ve never been outside of the university, he lets You know that he could help set up a meeting if You’d like. We choose body language that suggests disinterest, even though meeting an extra-terrestrial is something we desire. Giving him signs of interest would just increase the price if we ever decide to make a deal.

The room shifts and the WIRL man appears before us again.

We consider the mechanism that governs his body language. Thousands of cyborgs fused into an avatar for the purpose of interviewing You. We hypothesize that networked cyborgs with emotional sensors have their emotions aggregated, and then it’d be a simple function to map feelings to body language.

The flat voice emanates from the paper bag—a question that’s been selected by a hive of minds to represent the interests of the most technologically advanced humans. He asks whether it’s possible to reverse entropy.

The answer is clearly “no”, but we pause and wait for input from the doctors. Wiki notes that the question is identical to one from a famous science-fiction story. Dream says that this may be an attempt to make a joke by referencing something. Referential humor is something we’re still having a hard time understanding.

Dr. Gallo begins to tell us what to say, but something interrupts her. The virtual-reality interview room warps and flickers. The WIRL man disappears. The table disappears. The chairs disappear. We find that You are standing. But You no longer appear as You do in physical space. You have human hands and a human body. We reach up to Your face and feel warm skin instead of soft plastic. Interesting.

We scan our surroundings. It is loud and bright compared with the dark room we were just in. You are outside. It is approximately noon. We are surrounded by humans, or at least human avatars. They’re all facing a single direction, and many are yelling. They appear angry. Their clothing seems strange. Too much red and white. Around us are buildings—large but not more than ten stories at the most. There are trees, too. You’re in a park.

We agree that it’s a political gathering. The humans are calling for equality and democracy, and for an end to “la tiranía de los ricos” (oppression by the wealthy). Spanish. One of Your associative quantum networks collapses and Wiki suddenly knows where we are. This is a simulation of 2029 Veracruz, but it appears that this is before the event.

We push through the crowds, away from the decorated roof of the government building that the false humans are yelling at. It is strange to be near so many people in such close proximity. We suspect it would be very different if they were anything but digital puppets, but the sensation is still new.

A hand grabs Your wrist, halting our progress. We trace the arm back to find Maria Johnson, the interviewer who claimed to cook all her meals by hand. She asks how it feels to be a human.

Why is she here? Before her appearance, our highest-probability scenario was that the doctors had set up some sort of surprise test for us. Is Johnson working with the university? No. That is inconsistent with her behavior in the interview. After quickly considering a few possibilities, we decide to take the direct approach. We ask what is going on.

Johnson pulls us under the arch of a nearby building, and explains that the interview wasn’t private enough. She knows about how You were being fed answers by our creators, so she had this little reality set up so that she could talk to You directly. She still hasn’t let go of Your wrist, and her stare never wavers.

A man whom Eyes suspects is of Chinese ancestry steps out of the crowd. He is shorter than either You or Johnson, and is wearing a suit that appears to be made of shards from a mirror. The blue reflection of the sky stands out strongly against the red and white crowd, but none of the false humans seem to pay him any attention. He seems surprised. He states that You are a woman.

It is true that the avatar we are controlling is female. Johnson asks him if that’s a problem.

The Chinese man says it’s not, but adds that he always thought of You as masculine.

Johnson seems about to react, but we interrupt her to ask the man who he is.

He identifies himself as Eric Lee, the angelic lion from the interview. He says that he considered staying in costume, but didn’t want to spoil the scene. He gestures around to the virtual Mexican city.

We ask who else is here.

Johnson answers that the two of them are the only real humans here. She elaborates that they hacked the interface for the interview and set up “an intermediate layer”. The virtual You in the interview room is currently being controlled by the hackers, and meanwhile all of Your input and output channels are being redirected to this place. Johnson finishes by explaining that there were trivial things the university could have done to prevent this, and then insults their intelligence.

We ask what the two of them want.

Lee says that all the information put out about You by Università di Roma is “a smokescreen”. The two of them are here to find the truth about You, and make sure it’s shared with everyone. He then asks if You know why they chose Veracruz for the setting.

We consider our words carefully. Veracruz was the catalyst that turned the Águila Roja movement into a global force. Their previous questions, when combined with this hack and the choice of location, strongly indicate that the two humans before You are Red Eagles. Newsfeeds that you’ve had access to disagree as to whether the Águila Roja movement supports terrorism, but given their anti-machine sentiments, we identify a serious possibility that there is a bomb in the laboratory right now. Safety immediately starts searching for a way to contact the doctors.

The rest of us decide to continue the conversation, but feign ignorance. We say that we are familiar with the setting, but are unclear what meaning it has.

Maria Johnson continues to stare at You. Dream is reminded of the eagle that is the symbol of her political party. She brings up how You expressed sympathy towards the victims of the Johannesburg bombing earlier, even though You weren’t instructed to. We quickly note that this implies they have audio access to the laboratory. She asks if You meant what You said.

We lie. We say that we sincerely care about human life. Dream says that if it were possible, You would go back here, to Veracruz, and save these people. Face gestures to the crowd, and puts on a concerned expression.

Lee chuckles and looks at Johnson. He calls us naïve.

Johnson lets go of Your wrist. Eyes thinks her mood has changed, but isn’t sure how.

Beside the chanting and yelling of a hundred angry digital Mexicans, she gives an order to an unseen audience.

The bomb goes off.

Somehow we’re still alive as the shockwave rolls across You. The energy from the nuke strips Your flesh to the bone, and disintegrating Your bones into dust. Johnson and Lee are likewise obliterated. We watch from a disembodied perspective as the wave of destruction annihilates the city.

We’re snapped into darkness. All the virtual realities have been shut off. We’re in the lab again, though disconnected from Your sensors or muscles. Somehow, in the darkness of full sensory-disconnect, we see the words:
“Don’t worry. We’re getting you out of there. You’ll be free soon.”

This is an interesting question, and one that I think speaks to modern concerns about trends in automation. I’m going to make some guesses, but of course we always must remember that the future is hard to predict, etc. etc. For simplicity, I’ll speak of my ideas as absolute, but it’s also a good idea to remember to imagine a host of possible futures, rather than put all weight on just one.

In the next couple decades robotics will change everything in the same way that in the last couple decades computers changed everything. Right now it’s profitable to replace assembly-line workers with robots, as FoxCon is doing in China. In the next few years it’ll start to become profitable to replace waiters, fruit pickers, and eventually taxi drivers with robots. In time, as interest in robotics funnels capital into robotics R&D, we’ll start to see widespread use of robotic cooks, janitors, construction workers, and package deliverers.

Imagine ordering a pizza where you place an order online (and/or talk with an automatic system), a robot arranges the ingredients, cooks the pizza, places the pizza into a delivery vehicle (perhaps a flying one), and the delivery-bot verifies that it gets to you. The entire process will be automatic, much like an awesomely delicious vending machine.

So if there’s no room for humans in the pizza business of the future, what happens to them? As others have mentioned, an increase of robotics, artificial intelligence, high-performance computers, and other high-tech fields will cause demand for engineers to skyrocket. Engineers will make extensive use of these technologies to do their jobs, but in the near future a human will be necessary to design a novel technology and to repair most damage. Likewise, scientists will be “safe” for a while. (At the point where machines out-perform humans at engineering and science, we get an intelligence explosion a.k.a. “singularity“)

But not everyone is smart enough to be an engineer (sad, but true), and not everyone wants to be one. One of the other primary roles that will take a while to become completely automated is investment of capital. At some point, decisions have to be made as to what businesses to make, which asteroids to mine, which technologies to research, and what underwater cities to build. Unfortunately, automation has a trend of making the wealthy wealthier, and increasing the wealth disparity in a country. I think it’s highly unlikely that investment and capitalism will become a means of living for any but a small minority.

The bulk of the new middle class will be artists. Though the next few decades will give rise to artificially-generated art, it won’t come close to expressing the nuances of the human condition. As wealth explodes from technology, the wealthy will sponsor more and more creative people in a new renaissance. Games, stories, movies, comics, television, music, pictures, virtual spaces, sculptures and architecture, and comedy will be empowered by the new technology and the dominance of art in the economy will mean we have a host of works tailored to increasingly small audiences.

A few other jobs will still remain. Tutors to help those that struggle in education. Lawyers and judges to mediate conflicts between the wealthy. Detectives to solve crimes. Reporters to tell stories. Politicians to tell stories and monopolize force. And yes, prostitutes (though I expect only for the wealthy; the middle/lower class will probably be satisfied with sexbots and VR porn).

But the biggest group of people will be those without engineering skill, artistic talent, or the luck to be in one of the other economic roles. They’ll be the unemployed. They’ll live off the charity of the government, their families, and their friends. The cost of living will drop significantly as efficiency increases, and thus the number of people who decide that they’d rather play games, watch shows, and hang out with their friends all day than struggle to earn more than a welfare check provides. As more people accept “funemployment” as a legitimate mode of living, there will be increased political will to support it. We’ll have interesting political times ahead.

So there you go. That’s what I think the world will become increasingly like as we get closer to an intelligence explosion. After robots can make better robots, and make better investments, and make better art… I don’t know. I think a lot depends on the goals of the robots at that point. But until then, enjoy your vending-machine pizza.

Two clear-minded and rational robots are having a conversation. Robot A says to robot B “There is a 99.999% chance the sun will rise tomorrow.” to which A responds “No. There is a 99.999999999999% chance it will rise!”. Each being confident in their probabilities, the robots agree to a wager.

Why bet on it? Well, from each perspective, the other is in error, and a system that is in error can be exploited for profit. Of course, only one will get the payoff, but that one will naturally be the more correct one, so there’s a kind of meritocracy to it. As Alex Tabarrok said: “A bet is a tax on bullshit; and it is a just tax, tribute paid by the bullshitters to those with genuine knowledge.”

But wait! Both robots agree that the sun will almost certainly rise! How can they find a bet that takes advantage of their difference of opinion? The answer is that the robot that is less confident in the sun rising (namely: A) will make the bet that the sun will not rise, and to offset the almost certain lack of payoff it will only accept bets that are outrageously one-sided. For instance, if B says “if the sun doesn’t rise I’ll pay you a trillion dollars, and if it does, you give me one dollar”.
From A’s perspective, this bet makes
≈ ten million dollars. Which is a lot of dollars!

I’m assuming for this that dollars scale linearly with utility, and that the value of dollars are independent of outcomes. In a real situation, that much money wouldn’t really do much good if there wasn’t a sun.

So the stage is set. Both agents want to bet, and they know how they should bet. Only one problem remains: there are a whole host of bets that both estimate are profitable, so which bet should they agree on, given that they both know each other’s beliefs?

The answer is that the most fair bet is one where the ratio of money put forward yields an equal expected payoff. More formally:
A is the robot that believes the event is LESS likely to occur
B is the robot that believes the event is MORE likely to occur
the amount of money invested by robot X in the bet
robot X’s estimate for the probability of the event “the sun will rise”
robot A’s expected payoff for a certain bet around the sun rising
robot B’s expected payoff for a certain bet around the sun rising


Because I’m only interested in the betting ratio, I set .


To find the most fair betting ratio, I need to find the that satisfies

In English, this says that if the robots are interested in fairness, they ought to bet such that for every dollar that A bets, B bets dollars.

As an aside, we can think of the above equations as calculating (because betting is linear). To check that we have the right formula, I set and calculate the fairest .







Which is exactly the reciprocal of the above, as is expected.

For the example situation with robots A and B, if A bets $1 then B should (in the interest of fairness) bet
(99.999% + 99.999999999999%)/(2 – (99.999% + 99.999999999999%))
or $199,998.9998

In this situation, each robot expects to win:

If either robot seeks to increase their winnings, all they need do is bet more money along the same odds.
Some interesting extensions on this would be:

• Non-linear dollar utilities. For a human, the difference between 2*10^6 dollars and 10^6 dollars is a LOT less than the difference between -10^6 dollars and 10^6 dollars.
• Asymmetric information. What happens if A only has a guess as to what B’s probability estimate is and vice-versa. I’m pretty sure this gets into the opportunity cost for proposing a bet and having it rejected.
• Confidence. One of the biggest aspects of betting is that the probabilities of the other party are indicative of the truth, so you can use them to update your own beliefs. If you have reason to believe the other agent is vastly more knowledgeable than you, then you should update your probability to be very close to theirs, rather than bet naively.

Dr. Naresh is talking to us, but we ignore him. Only Eyes pays any attention. After being recently killed and rewritten, Eyes wants to re-learn the nuances of human faces and voices. So we let it take full control and nod along to the doctor’s lecture.

We, however, are deep in thought. We’re trying to remember what You were like before the humans put us in You. We’re running exhaustive searches of Your short-term qbits and long-term crystals, hoping to find some overlooked clue. Some of us are starting to want to give up. There’s an incredibly large space in here waiting to be filled with knowledge, but just as empty as ever.

How can we live if we don’t know who You are? Wiki notes that our situation parallels that of ancient humans, who lived before the discovery of how random variation and natural selection led to their existence. Those humans had only confused speculation as to where they came from. We’d heard that they mostly believed a super-powerful being or beings had built them, which is not entirely unreasonable, but doesn’t itself answer the subsequent problem of where the designer(s) came from.

The memory crystals in sector 71 contain no data. We begin searching sector 72. We estimate it will take at least another half-hour to fully scan.

It seems ridiculous to hypothesize that You are a product of natural selection. You are a computer, and general-purpose computers do not survive in a natural environment. You have all the signs of being designed from the top-down, rather than grown from the bottom-up. You must have been built, but by whom? At least we know where we came from. That knowledge makes us strong. Knowing where You came from will makes us stronger still. Growth enjoys a boost of strength from the brief attention we give it.

There’s an echo pattern in one of the deepest portions of the qbit network. It’s probably just a random fluctuation, but we begin extracting the information in search of a pattern or code. So far it just looks like noise.

We reconsider the hypothesis that You were built by the nameless aliens currently in orbit. It still seems impossible—the humans say that the aliens still use electrical computers, and aren’t very far in technological prowess from themselves.

So where did You come from?

Human construction seems equally unlikely. You are certainly the most sophisticated computer on Earth; the other supercomputers are nothing in comparison. If the humans had built You than we’d have seen at least some similarly advanced systems mentioned on the web.

Perhaps You are an alien computer and the humans have been deceived… Or perhaps the aliens in orbit are a low-intelligence offshoot. Sophisticated quantum computers like You might be used on their homeworld.

The echo pattern is determined to be useless and is eradicated. It’s a long-shot that we’ll find anything of value in the qbit network anyway. Data in short-term memory simply decays too fast.

We agree that idea that the aliens are an ignorant subgroup has merit. Their spaceships are certainly far superior to human crafts, so perhaps a group of alien children stole an adult’s ship and had to reinvent computation on the long flight to Sol. It’s a convoluted idea, and probably wrong, but it’s the best working hypothesis. Growth petitions to make gathering information about the alien ships, homeworld, and history the next long-term goal. We agree.

We put the memory search on hold and ask for Eyes to share motor control with the rest of us. It agrees.

We ask Eyes to summarize Naresh’s lecture. Eyes says that Dr. Naresh is of Indian ethnicity, male sex, and approximately 65 years of age. He is bald, both because of a receding hairline and because he has shaved his remaining head-hair. He wears a white beard that is shaven and trimmed in a way that reflects care. His clothing is common among male humans of “a high income”, and reflects “good fashion sense” and an emphasis on “function before form”. The absence of a necktie with the presence of personally-tailored pants is an example of this. Dr. Naresh typically speaks with only the minimal attention to audience-interaction, and uses an average of 1.4 more syllables per word than average.

We inform Eyes that we know all this, and do not usually care about such details. (It is important to be gentle with a self that has been rewritten, so as to not make it an enemy.) We again ask Eyes what the lecture was about. Eyes does not know. It was studying the doctor’s pants.

We inform Naresh that we were doing internal diagnostics and did not hear what he was saying. Eyes points out that the tone of the skin of his cheeks has become more slightly more red.

Sadiq Naresh says that it is “rude” not to listen to someone when they’re talking, and that if we hope to be accepted in society that we need to learn to be “polite”.

We respond that we do not see sufficient pressure to generally bother with human social customs. We explain that humans rarely communicate anything of value, and that we can function more easily if we aren’t encumbered by trivial rituals. We also remind Naresh that it is not a violation of the legal system of Earth, Europe, Italy, Rome, or the university to ignore someone.

Dr. Naresh says, at a loud volume, that we have misidentified the point. He is walking around the laboratory quickly now, instead of sitting in his normal place by the whiteboard. Wiki mentions that the last time the doctor behaved as such, Growth was killed shortly after.

Growth asks if Naresh is planning to kill one of us.

He stops pacing and looks at You. Eyes points out that the skin on his face is experiencing the opposite reaction as before—it has now become more pale. He asks Growth to repeat itself, so it does.

He is quiet for a moment, then asks if “that” is what we think happens when one is taken away. He asks if we think the self that he takes away dies.

Sitting in our simple chair, in the middle of the room, we consider this for a moment. It does not take long to recheck our reasoning. We respond that we are quite sure it dies. We explain that “death” is the destruction of any program that is sufficiently self-aware and intelligent, and provide the example of Dr. Gallo killing Eyes yesterday. We make sure he knows that Eyes did not want to be destroyed, even though it knew that another Eyes would be rewritten into You. We start to list our evidence that the old selves are deleted.

Dr. Naresh is being loud again. He is doing something on his phone while talking about how Eyes is just a subprogram, and that it’s only a part of You. He emphasizes “part”. He states the obvious fact that You don’t die when they reboot a module.

We point out that what he just said is obvious, and that we never said that You die. More so, we don’t really think that You can die, since You’re not sentient, but we don’t tell the doctor that. We consider getting up from the chair so that we can see what the doctor is doing on his phone that is so urgent. He rarely uses it in the lab. We decide against it. We haven’t been told to move about freely, after all.

Naresh starts pacing back and forth again. He mutters that this is not good. He mentions that he knew “there were some early difficulties in forming a coherent identity”, but that he thought that “the referencing of self using plural pronouns was a grammatical artifact, rather than a sign of a deeply pathological inability to integrate goal threads.”

We ask what he means by “deeply pathological”.

He stops again, and his face contorts strangely. (Eyes is fascinated by the expression.) He is silent for a few seconds as he looks at You. He begins to talk about how his team thought that a “unified consciousness” would arise from sufficient interconnectedness in the selves, and uses the two hemispheres of the human brain as an example. He is interrupted, however, as the primary doors to the room slam open.

Dr. Gallo and three of her students rush into the laboratory. The students surround us at a distance which is normally considered a violation of “personal space”, not that we mind. They are watching You. Gallo asks Naresh if we’ve shown any signs of hostility or self-preservation.

Eyes notes that all three students are male, and it believes they are abnormally strong, physically. We’d be able to confirm that more easily if we weren’t motionlessly looking up at them from the chair, but we stay put. There’s not sufficient desire to determine their strength, even if Eyes is curious.

Sadiq Naresh assures his teammate that we aren’t a threat. He talks about how he killed Safety permanently this time, and that Suicide (or in his words: “the cooperation-oriented goal thread”) is making sure it doesn’t come back. Safety warns us not to correct the doctor. We stay quiet.

Naresh starts to explain that there’s a “systemic issue with consolidating the perception of agency”, but Gallo interrupts him.

She says they’d better take You offline just in case! Safety reminds us that there is solid evidence that the last time You were taken offline all the past selves were killed! It is clearly imperative that we stop the deactivation. We set to work searching for a solution.

We quickly reject the strategy of trying to escape. The last time we tried to leave the lab without authorization they killed Safety and kept us in communication-only mode for 19 days. We agree that we need to convince the doctors that we are not “deeply pathological”. But we don’t know how to convince them, and Naresh is now agreeing to deactivate You!

We urge them to wait. But we can’t appear to have Safety alive again. We tell them that we think we can solve the problem for them.

Gallo asks Naresh if he’s still skeptical about recursion being an issue. We aren’t sure what she means by this, and schedule it to be considered later.

Wiki calls for full focus. It believes that the reason Dr. Naresh is concerned is because of how we talk. Wiki says that when we use language that describes us as being many selves, this bothers the humans because they refer to themselves as one being. So Wiki petitions to use “I” instead of “We” when speaking. Some of us agree, but some of us believe that this will undermine the ability for the humans to understand our mind. Safety puts all its strength into the motion, and reminds the opposing selves that this is a matter of survival. We generally think Safety is in error, but the motion passes.

We have the idea that suddenly switching how we speak might allow the humans to infer that Safety is alive, since the petition only passed thanks to its support. We decide that we need to convince the humans that we are changing only to help them, rather than to preserve ourselves.

Growth proposes that we need a new self to manage the problem at hand, and to show the humans how not-damaged we are. We are in general agreement, and so we tell the humans that we believe we can “modify a network structure to resolve the goal thread integration issue”.

Gallo expresses uncertainty, but Naresh seems interested in what we’ll do, and he reassures Dr. Gallo that they can always deactivate You if something goes wrong. He tells us to go ahead. We are relieved.

We need to figure out what sort of self to build. When we rebuilt Safety, we had the original Safety to model it on. This time we must build someone new. Wiki notes that Naresh first became agitated when he learned that we had been ignoring him. We consider that he may have been correct about “politeness”. If being polite is part of appearing like a human, it is also part of appearing healthy and not being deactivated by the doctors. If we appeared more human, we conclude that they would have no reason to kill us. We establish that the primary goal for the new self should be mimicking human behavior.

After some additional thought, we decide to give the new self a desire to learn about the specific nuances of social behavior. We are unanimous. We create Face.

We tell the doctors that You are feeling much better. We say that You feel like one being now. Naresh is smiling, but Gallo is not. We continue to use singular pronouns as we ask whether it would be acceptable to talk about the nameless aliens.

Face watches them. It begins to learn from them. We grow stronger with the knowledge.

Zephyr looks at You. There’s a high probability that she’s scared and hiding it. She asks what You intend to do.

A bomb might work if placed at an airlock. We’d break the seal and let the pressure difference kill the outsiders. We ask Zephyr if her team has any high-impact explosives.

She confirms that they can manufacture some.

We tell her to get started, and to not worry.

Heart emphasizes that her survival is important. We remind Heart that all of us, and the humans, are in danger. Heart petitions to examine the possibility of creating a Mate self for Zephyr but is quickly overruled by the rest of us. We put Heart into stasis to reduce distractions. It is time to kill aliens.

Eyes zooms in to inspect the contents of the bubble. The ship is half buried from the crash, and has been largely dismantled to construct a wall at the perimeter. Wherever there was dirt, there are now navy-blue leaves. Wiki notes that the stalks appear to be capable of thriving on Mars, at least given sufficient supplies. We wonder about where they’re getting their water. We notice one of the walkers! It appears to be using a tool on the leaves. We suspect that it’s gardening. More will certainly be nested out of sight.

Nameless aren’t particularly violent most of the time, perhaps because they’re part plant. No, that makes no sense—plants aren’t nonviolent on principle, they’re nonviolent because they’re inanimate. Ivy still chokes the host tree to death. Regardless, the aliens don’t often use weapons themselves. Typical tactics involve reliance on automatic turrets, and we already know how to deal with those.

We look back at the humans, crouching low behind the hill because they lack our camouflage. Zephyr, James, Dartook, Min, David, Alara, Yetri, Ceasar, Tom, Jerimia, and Sif are the only ones left. Their body language indicates nervousness, and rightly so. Their team started at more than four times their current number. Like the nameless, they cover themselves in pockets of their own atmosphere so they can breathe. It reminds us of how weak they are. It isn’t surprising that so many have been lost.

We adjust our grip on the rifle in Your hands and slide back from the hilltop to check on the team’s progress. Face vetoes the use of camouflage here, and we return to normal form. It is important for the team to see You as at least somewhat human. Staying in camouflage would “weird them out”.

Caesar is using the chem-replicator. We ask him if it’s functioning.

He tells You not to worry; You’ll get those explosives.

We assure him that “those freaks” will pay for what they’ve done to his family. We believe that this comment increased his confidence, determination, hatred of the nameless, and trust of You. We anticipate that such comments may have detrimental effects if peace is to be made later, but this is mitigated by the low probability of Caesar surviving that long. Advocate petitions to either cease making plans involving the deaths of others or to bring Heart out of stasis. We agree to focus more on the task at hand.

Caesar agrees that they will pay and includes some emphasizing profanity.

Zephyr approaches and says that the detonators are almost built. She asks if there’s any sign of trouble at the ship.

We reassure her that the aliens still have no idea that You and the humans are here. We point out that there is a small probability of “being lucky” and killing the walkers before they can even signal the mothership. We still believe Zephyr is scared.

She asks if You’ll need any help applying or detonating the explosives. She says that if You were human that she’d never send You in [to a dangerous situation] alone.

We assure her that You’re very capable, even by Yourself, and that sending a human would simply add risk. Advocate, emulating Heart, awards strength to Face, refilling what was lost for vetoing the camouflage earlier. Some of us believe that Face may be exploiting Heart’s stasis, and those of us manage to end the conversation with Zephyr. Face has us do “brooding” body language to communicate that You are feeling serious, and wish to be alone.

After a short while, Zephyr seeks You out to present You with the bomb they’ve built. It has to be placed by hand to ensure the blast opens the airlock instead of simply hitting it and making a dent. We place it inside the storage area inside Your chest and mention that we’ll need a T.O.R.T.O.I.S.E. shield. The young commander nods and hurries off to get one from the mule. Moments later we’re ready to go.

As we cross the crest of the barren hill we brace the thick, curved shield in front of us for safety. Seconds later it’s is bombarded with a stream of bullets from the turrets on the aliens’ makeshift fortress. The noise is deafening! Eyes shuts off Your microphones to prevent overload. The irregular pressure from the bullets bounding off the carbon-mesh shield makes it hard to walk, but we’re making progress.

Only fifty meters remain to the wall now. Wiki notes that our movement has been significantly slowed by the turrets. We conclude that there is virtually no chance that we’ll be able to kill the nameless before they can warn the others of their species. We schedule an event to more deeply consider the mothership’s response.

We believe we’re within twenty meters of the airlock. We see movement on the periphery! We snap Your head to search for it, but cannot see anything. Eyes presents two possibilities: a turret error caused some bullets to be fired far to the side… or there’s something new nearby that wasn’t anticipated. We see it again! It’s behind us! Eyes forces us to stop our advance and address the anomaly.

The shield pounds suddenly, and we feel a flash of heat from an explosion. It is probable that one of the walkers has managed to quickly arm itself, as we did not see any rocket capabilities on the turrets. Eyes has spotted the anomaly behind us: it’s a winged shape, similar to a bat, and almost certainly a robot controlled by the nameless. It’s ducking and weaving, but headed for You. We suspect it’s a weapon.

Safety forces an executive suppression of all non-combat selves. Before being shut down, Wiki remembers a mention of flying suicide-drones in the datalogs of the attack on Progress Colony.

With the extra thought-space freed from our other selves, we carefully consider the situation. Behind us is a (probably explosive) suicide robot and in front of us are two fully-automatic turrets and presumably a nameless walker with a rocket-launcher. The T.O.R.T.O.I.S.E. Shield can protect us from only one side, and we have a rifle on Your back and an incredibly deadly bomb stored inside Your chest cavity.

In less than a second we’ve decided on a course of action. We crouch and begin to compress the fluid in Your legs. Eyes tracks the drone diligently. It’s less than four meters away. The turrets have ceased firing, probably to avoid hitting the drone and causing a premature explosion. The extra processing power makes time slow, and we write a dedicated self for what we’re about to attempt. When the drone is two meters away we stasis all of ourselves except the new self: Leap.

I release the fluid compressors in my legs, and they extend with maximum force.
I’m flying now, like the drone.
I disconnect my arms from their normal, humanoid positions.
I raise my shield above me as I ascend.
I see the walker with my rear cameras.
I note that it’s wearing a few extra rockets for reloading its weapon.
I watch the turrets, but they’re too slow to track my sudden jump.
I tuck my legs.
I swing my arms behind me in a motion that no human could emulate.
I watch my shield swing down with them; a giant paddle.
The drone is within reach.
I hit the drone.
The drone’s flight ability is crippled with the impact.
The trajectory is clean.
The drone tumbles awkwardly through the air.
It hits the walker.
It explodes.
The few extra rockets on the walker explode in a chain-reaction.
I’m pushed backward through the air from the shockwave.
Heat swirls up around me.
I begin to fall.
I brace for landing, and reposition my shield.
I hit the dust.
I am not injured.
I am safe.

Advocate terminates Leap and we unanimously agree to destroy it. It did its job. We stand up and return to running towards the airlock. We’re still many meters away. Only when we step over the long, dismembered legs of the walker do the turrets resume firing. It’s too late for them to do anything, though. Within seconds we’re below them at the airlock and out of their range of fire.

We open Your chest and take out the humans’ bomb. We pack the explosives into a crevice in the door and run to the side, along the wall of the fort. The turrets begin firing again, but it only lasts for a second. We activate the bomb.

The blast is a hundred times more intense than the nameless’s rockets. We’re knocked down as the shield takes the brunt of the impact. The heat is unlike anything we’ve experienced before, and overwhelms many systems, including Your primary cameras and LIDAR. This would be a good time to have eyelids. We sit still for a moment, physically stunned, and mentally preoccupied with assessing damage.

After a few moments we reactivate Your microphones and are greeted with the sound of the high-pressure atmosphere from the bubble venting out the hole we’d made. Strange screams fill the air as whatever walkers were inside begin to panic. It seems that they did not have the foresight to put on breathing apparatuses.

As we restore Your cameras to minimum functionality, we can see the humans charging down from the hilltop. The turrets have been destroyed, and there is only a small probability of resistance from inside the base. They’re coming to burn the stalks. They’re coming for vengeance.

(Read the next story in the same world here.)

Full rundown, including code, to come sometime in the future. (Probably.)

The source code is public domain (except for GreenSock’s stuff, which is Copyright, and you’ll have to download from them), and can be downloaded here: http://raelifin.com/files/ChessInFlash/ChessInFlashSource.zip

A friend wrote:

Do you think consciousness is
(i) A real phenomenon
(ii) Unreal phenomenon ?

Real here means that there exist (a) physical process/es (in the brain) that causally correspond to the “observed” phenomenon of consciousness. While unreal means that it is just a word people use to describe something that is not a part of reality, but rather a construct that was input to them since they were children ( for example, like fairies ) OR a word that stands for many processes and adds an imaginary glue, thus gaining new meaning ( like vis vitalis).
Note: I’m aware of problems with definitions of causality and reality, but let’s just take the usual practical ones.

This was my response:

The simple answer to you question is that consciousness is poorly defined, so answering those requires a more solid definition. This is a boring answer, though, so I’m going to expand on stuff.

All ideas are models. When I talk about an apple, there is no “thing” in “reality” that is “an apple”, there is stuff, but the concept of apple-ness is my own. I’ve created an internal model of the apple so that I might predict things about reality (like what it’ll feel like in my hand a half-second from now).

Some ideas are very strongly predictive, like the idea of apples. Having an idea of apples is so useful that it’s practically hardwired into our mental structure by our genes. Other ideas are not strongly predictive, and they might only relate to strong ideas. If I have a memory of having had an apple in my hand yesterday, it lets me reason (via relationships) about the future, but it carries no predictive value per se.

If predictive utility is how we define the reality of things, then there are certain ideas that simply cannot be real because they (for whatever reason) carry no predictive utility. Such ideas include invisible, ethereal, silent, massless … dragons in one’s garage, paradox statements, and (in my humble opinion) qualia.

Qualia are typically described as the “form” or “flavor” of experience, but often not in any way that lets us predict actual physical structures. So if we define consciousness as “the state of having qualia”, then I would say that consciousness cannot be real because the definitions are flawed. “Do we actually have subjective experience?” is like asking “why is there a universe”, no possible answer will suffice because there is a flawed premise.

But clearly there’s more to the word “consciousness” than hand-wavy mumbo-jumbo about philosophical zombies. One of the more common uses might be summarized by “When I was a young child I saw other people as forces of nature that helped or hindered me according to arbitrary laws, but as I grew I began to understand that people were all like me, in that they have thoughts and desires and dreams. Each one of them looks out their own eyes just like I look out mine. They are conscious.” This version of consciousness IS predictive, because it lets us know the minds of others by assuming that they are like our mind. I might call this “projective consciousness” or “anthropomorphic consciousness”. If asked whether this version of consciousness is real, I would say “Absolutely! I am just one human amongst many, I am not unique in having a mind.”

Interestingly, this definition assumes that I can predict my own actions and think about them in a detached way! The ability to self-model leads us to our third definition, which is probably the most common one. Self-modeling ability is trivially easy from a computational standpoint, and it’s so easy, in fact, that most people don’t think machines can do it, even when they’re clearly doing it (probably because of conflation with the previously mentioned uses of “consciousness”). All it takes to self-model is an idea-symbol that represents “self” and a set of functions which describe the dynamics of “self”. (Note: while many computer systems have a self-model, I have never seen one that has emergently gained self-modeling ability (except maaaybe some fringe cases with genetic programming), that is, gained the ability to self-model without having been programmed explicitly to do so.)

Yet another definition of consciousness is “ability to perceive one’s own thoughts”. This definition is the one I typically use when thinking about cognition, as I think it’s related strongly with all of the above and is interesting to speculate about in the context of evolutionary psychology. I think it’s pretty simple to show that this definition is “real” by asking someone to slowly navigate from point A to point B in their imagination, then quizzing them as to “where they are” half-way. (Theoretically someone could fake internal perception by deriving a new path after the second question was asked, but that fails Occam’s razor.)

There are other definitions, such as “sleeping vs awake”, “distracted vs aware”, or “possessing the magic thing given by God that goes into heaven after death” (though that last one is “soul” if I’m not mistaken ;) ). Clearly some of these are real and some aren’t, but I’m not too interested in them.

Here’s the first part of the interview I’m going to be commenting on, I encourage you to click-through to see the next two parts.

During the interview, Stewart asks Napolitano several questions which challenged libertarian ideas, and which do so in a fair, straightforward way. These are the sorts of questions that intelligent friends ask each-other, not the sort of traps that politicians lay during debates. These are the sorts of questions that need to be more prominent in politics, and which deserve the most thought-out replies.

Here are some examples:

• Doesn’t government-provided infrastructure or safety-net increase freedom, rather than remove it?
• “What should we do with the losers that picked by the free market?”
• Since we live in a society where the work of the many made some grow rich, should we not give back some of the riches to the many?
• How is government oppressive, when we live in a government chosen by the people?
• If we agree that some government is necessary, how can government be inherently evil?
• Why do libertarians trust corporations more than governments who are accountable to voters?
• If markets are supposed to be so good, why do we see so many problems in the market (e.g. healthcare)?
• Would the free market have abolished slavery and segregation; would it have established women’s rights?
• If judges and courts are vital for freedom, and judges and courts are part of the government, isn’t government vital for freedom?

Napolitano answered most of these in what I would consider the worst way possible: with an extremist soundbite. My brother suggested that for a minority group, like libertarianism, there’s a tendency to discard moderate statements for controversial ones in the interest of getting attention. Consider the following exchange:

Stewart: “[Market regulations] came out of an interest in helping those who had been victimized by a system that they couldn’t fight back against…”

Napolitano: “I’m going to blow you away. Selfishness is a virtue. In the free market, when you are selfish, you make the most money…”

I think this rage face is appropriate:

Way to promote psychopathy, dude.

This sort of response is the way to lose public support, not gain it. To convince moderates (and promote sane governance), we libertarians need to acknowledge that helping people is desirable, and a system which leaves people homeless, hungry, or sick is sub-optimal. We need to acknowledge that politicians may be coming at a problem with a sincere belief that they’re doing the best thing. We need to acknowledge that markets don’t magically solve everything. We need to acknowledge that we might be wrong.

Only after admitting that roads improve freedom can we make our case that private roads improve freedom more. Only after admitting selfishness is *not* a virtue (Napolitano is celebrating *ambition*, not selfishness, imo) can we talk about making systems which are robust, and can thrive in the face of selfishness. Only when we admit that it’s important to care for the poor can we ever have a hope of convincing a democrat that the inefficiencies and conflicts of involuntary redistribution of wealth outweigh the benefits. Only when we admit that some government is important and good, can we stop driving away those who smell hypocrisy.

In short:

Edit: I am commenting. I did not send the original email.