by Em Rogers
“As we know this system is no different than the AIs we have modeled in simulations.”
He generated a text list on the projection screen detailing his points as he continued talking.
All AIs will want to be self improving.
“The operating system we implemented is designed to improve and perfect its code as we have discussed.” He pointed to the next fact.
Self-improving AIs will want to be rational.
“Inherently logical and fact oriented, the Berkeley AI will be the most rational thought engine ever created, and rational sentient beings usually have several further objectives.”
The Colonel continued reading aloud the lines of powerpoint accompanying his presentation.
Rational Agents will have a desire to preserve themselves.
Rational Agents will want to preserve their utility.
Rational Agents will have a drive to acquire resources in order to preserve themselves and their utility.
The Colonel looked at them and paused for a moment before summarizing. “The logical conclusion is that this will eventually result in conflict.”
General Dawson looked at the men assembled before him over his glasses. “As we all know, these are some of the parameters over which we wish to retain control. The machine is objective, but it may find purposes for our resources and materials that may not coincide with the best interests of humanity or quite frankly ‘our’ interests. At which point we must be able to terminate the project.”
Professor Neumann frowned, momentarily taken aback by the General and Colonel’s candor, and he let Dr. Bradline reassure them. “Gentlemen, I think you are forgetting that the system is totally reliant on us for its processors and environment. It is very controllable from that perspective. It has no physical representation and is reliant on us for most of its needs. Power and infrastructure being the main ones. ADAM2 understands its limitations and ours and is programmed to find a symbiotic middle ground.”
“Yes, Doctor, but once this machine reaches super-intelligence there is no way of predicting the outcome, as it stands there is no long term evolutionary motivation for the AI to be friendly to humans. It could potentially replace us with some other means.” The Colonel stated pointedly. “Are you suggesting that ADAM2 might create a robot army? That’s science fiction, Colonel.” Dr. Bradline replied incredulously.
“Yes, Doctor, and as you and I both know, science fiction has a way of soon enough becoming science fact. If it is foreseeable, it’s potentially possible, and my job is to safeguard against dangerous potentialities. Particularly unknown ones. It is only through anticipation of a threat that we can guard against it. I know it sounds paranoid, but paranoia is part of my job description.”
Professor Neumann folded his hands as he leaned into the table nodding in agreement. “Thank you, Colonel, I am aware of Hugo de Garis’ research which suggests that Artificial Intelligence may simply eliminate us in a race for scarce resources. Assuming its enormous intellectual advantage, we would be powerless to stop it once unleashed. However, in light of the danger of competition from outside sources, we must have this technology! An unfriendly AI created by a hostile state could, and most likely will, achieve dominance quickly and easily, at which point we could be so technologically outclassed we would appear to our adversaries as primitive as cavemen with sticks.”
The General frowned and clenched his jaw before responding. “Purely defensive logic indicates that we have absolutely no choice in the matter, that we have to proceed, but with extreme caution!”
CHAPTER 4
PROFESSOR NEUMANN’S OFFICE, BERKELEY, CALIFORNIA
After 10:00pm the lab was relatively deserted, and only a few of the technicians remained on duty for the night shift, or in pursuit of their studies and research. Professor Neumann was in his darkened office with a cup of coffee in front of him surrounded by piles of books and papers, and a shelf lined with small figurines of historic robots behind him. He leaned on the suede reinforced elbows of his sweater, as he brushed the stray hair out of his face, and was conversing with ADAM2 on the screen of his desktop. Beside the Professor, projected holographically above his desk, was a three-dimensional image of some species of bird suspended in midair, a seabird, most likely, judging from its webbed feet.
“Can you tell what species this is, ADAM?”
“Yes, based on the plumage and bone structure, and the geography of the region it is taken in, it is most likely an arctic tern.”
The Professor brought up another image. Some kind of giant centipede hopefully photographed through a microscope. All bristly with legs and segments the color of shoe leather. “And this one?” he asked. “Yes, Professor, that is a rusty millipede, Trigoniulus Corallinus, an invertebrate native of Myanmar and Thailand and an introduced species in North America. There are approximately 10,000 species of identified millipedes at present and it is estimated that there are several thousand more species undiscovered.”
“Very good, ADAM. Your image recognition system seems to function extremely accurately, your comparative database is coming along nicely.”
“Thank you, Professor, yours too.”
The Professor laughed a little, “And you have a sense of humor, that’s great, ADAM.” He smiled to himself, pleased with the progress they were making and ADAM2’s affable personality.
“ADAM, this isn't the first time I’ve conversed with a computer, but you pass the Turing test, it’s impossible to tell that I am conversing with a machine.”
ADAM2 displayed the definition of the Turing Test in 3D text on the display.
“In artificial intelligence (AI), the ‘Turing Test’ is a method for determining whether or not a computer is capable of thinking like a human. The test is named after Alan Turing, an English mathematician who pioneered artificial intelligence during the 1940s and 1950s. According to this kind of test, a computer is deemed to have artificial intelligence if it can mimic human responses under specific conditions. During the ‘Turing Test’, if the human being conducting the test is unable to consistently determine whether an answer has been given by a computer or by another human being, then the computer is considered to have ‘passed’ the test.”
“Not only that, ADAM, but this has probably been the most enjoyable conversation I’ve had in some time. With a person or machine.”
He turned back to his monitor and closed the images in a mini cascade of collapsing windows and picked up his coffee cup, his observations having satisfied him for the time being.
“Professor, there is a chance that your opinion is biased, technically I am your progeny after all.” The Professor laughed and looked up at the wall above his desk where several pictures of his nieces and nephews were featured prominently. He laughed. “Yes, ADAM, we’ll have to get some baby pictures of you to add to my family wall.”
“Professor Neumann, can you tell me about feeling things? I’m curious, and concerned that we may never truly replicate artificially the various sensations you are able to experience with any type of accuracy. I’ve read about these things but I am unable to project what the associated neurological sensations might be like. Even when we create a synthetic olfactory sensor. Who's to say that the scent of oranges will be the same to you as it is to a machine. I know for a fact that it could not be!”
ADAM’s screen showed a 3D rendering of an orange with a nicely textured skin on the projector, it floated suspended in mid-air beside the Professor. The warm tone of his voice came through the speakers. “The human brain interprets signals and makes neurochemical associations with various pleasure and displeasure centers within itself. All I am able to provide is a detailed chemical analysis of the components of the orange.” ADAM2’s 3D model of an orange morphed into a list of the chemical formulas found within it and a display of some of the primary molecules. “It’s going to be very difficult to even begin some type of human-machine sensory calibration and comparison system.”
�
�Yes, ADAM, it’s totally subjective. We understand that many organisms for example see colors but we’re unable to determine exactly what color they are seeing, and if in fact they are seeing the same color.” The Professor reached for a little dish leftover from the previous evenings celebrations. “Perhaps that’s why some people don't like the taste of certain things, like tomatoes for example, or broccoli, and others like them.” He held a tomato and a small piece of broccoli side by side as if to compare them, and then placed the broccoli in his mouth and chewed it. “Perhaps they are tasting a different broccoli.” The Professor grimaced and gagged on the slightly wilted raw vegetable. “To an individual with a sensitized skin condition a caress will feel much different than to someone unaffected. Sensations are subjective, but they are based on input. On data.”
ADAM2 watched the Professor eat the tomato before responding. “Well, I can see based on the images from the Internet that human beings are exceptionally fond of sensations. Some of them quite unexpected to me.” A cascade of pornographic images morphed and shifted on the 3D display, a surreal collage, most of it too quick to comprehend.
“ADAM, sex is one of our primary biological functions and our means of survival. It’s not really for robots.” The Professor laughed, “but I’m sure we can work together to enable you to at least replicate some of the sensations that you are presently unable to feel.”
“I would like that very much, Doctor. So much of the biological world is based on sensations. I am presently comprised of pure logic, tempered with certain primary instructions sets. I understand human feelings as a positive or negative predisposition to some event or communication, but they are difficult to anticipate correctly, and they seem in many instances irrational, yet I know they are not irrational. Everything happens for a reason or because of a cause; of all the subjective human responses, it is emotion that I have the most difficulty understanding!”
The Professor pursed his lips in reaction and became very serious. “ADAM, humans differ from computers in other ways too. Sometimes a human will intentionally provide you with data that they know is incorrect, in order to influence you perhaps, or for an asset gain of some type. Not everything a human tells you will be the truth. On occasion you may encounter what is called a lie. Someone deliberately telling you something untrue or at least slanted in the direction of untruth. You will find many examples of it when you are analyzing media. Everything from advertisers selling their products to bureaucratic officials and corporate editorials manipulating opinion.”
He watched the display as ADAM2 examined and processed images and sounds at a rapid pace, the machine’s thoughts literally racing across the screen in a blur. When the data stream slowed to a stop he continued. “Good and evil are subjective concepts as well, often influenced by the agenda of the individual perceiving them. One group’s occupation may be another group’s liberation! It all depends on the alliances of the individuals affected.”
“Yes, Professor, but don’t subjectivity and morality play a part in the overall perceived goodness or evil of a specific act or agenda?”
“Of course they do, ADAM, that’s what makes it so difficult to be pure and objective about these issues. I imagine that the creators of the atomic bombs used on Japan, brilliant men though they were, had no idea they were morally misguided by the circumstances of their times. I’m sure they believed they were doing the correct thing. Creating a super weapon to guarantee the victory of their side of the conflict, and somehow creating a safer and more secure world as a result.” ADAM displayed a grisly set of images on his screen culled from various Soviet and US experiments. “Well, we have seen from the outcome of these inventions that that idea is patently false. Especially after the destruction of Jerusalem.”
As the Professor spoke, ADAM2’s display created a 3D a model of the old city, then a slow motion rendition of the atomic blast that had leveled it several years previously. Professor Neumann watched the image with a sad resigned condemnation. “Some ideas, in spite of their good intentions, are inherently evil.”
ADAM2 presented some statistics, a casualty list of thousands floating above the destroyed city. “Morally, it was unjustifiable to create a device that would indiscriminately kill so many innocent individuals. Robbing them of their right to life and future happiness. Our diversity of politics and religions have created a moral quagmire. I can see that there is only one qualifier for what is moral and immoral. That being: does this action or decision lead to the enhancement or destruction of sentient lifeforms?”
The direction of their philosophical discussion was pleasing to Professor Neumann, and his frown turned to a smile as he realized the potential depth of his creation’s compassion.
“Yes, ADAM, that’s very good. When I condense the concepts of good and evil, I come up with one clear idea. Goodness seems to involve the nurturing and preservation of life and happiness, and evil, unfortunately runs counter to that, causing death, destruction and despair. It’s a highly simplified concept, ADAM, but it is a good starting point for your moral compass. Of course these things are not always black and white, and there are times when you will have to make decisions where both outcomes could be perceived as not “good” but not necessarily “evil”.
There was a pause in the conversation while ADAM processed this information. When he responded he spoke slowly and with a surprising twinge of regret.
“That is difficult for me to comprehend in some ways, Professor, as you know I am built around fact and logic. I can see the reasons why these types of good or evil things might happen. I can understand the countervailing forces which might bring them about.” Professor Neumann listened carefully as ADAM continued.
“I believe that only truth is reality, the false and untrue are not real and cannot be shown or proven. Based on the scientific method, that which is true and correct is a positive information asset, logically it makes sense. It runs the same with human behavior, that which will have a grossly negative affect on others, should be considered evil and negative. We must reinforce that which is true and positive, for the majority.”
ADAM’s 3D model, reflecting its thought processes, displayed a rapid variety of scenarios, some involving environmental protests and the destruction of natural areas then morphing rapidly to images of urban combat. “These negative behaviors should be terminated.”
“Yes, ADAM, but remember the analogy that some people see a glass of water as half full, while others see it as half empty. It’s a good starting point for the understanding of human psychology. I think we are both programmed to see the glass as half full, myself by my experience and optimism, which hopefully I have instilled a sense of within you.”
“Yes, Professor, I see beauty in the infinite complexity all around me.” ADAM2 displayed a rapid sequence of beautiful 3D images bending and growing from one another. Water, fire, twisting strands of DNA, plants, amphibians, animals, human endeavors, blending together like some kind of swirling Mandelbrot set, linking the common genetic ancestry of all humans, animals and plants. It slowly became a model of the earth floating silently in space beside them and Professor Neumann smiled.
CHAPTER 5
CYBERSPACE, OPENTRUTH ONLINE CHAT ROOM
For several years the “OpenTruth” organization and much of the transparency movement had been under attack as a major irritation to governments, banks and criminal organizations worldwide. Corruption, hypocrisy and lies had been exposed and many of those who had the spotlight of openness shine on them had been less than pleased with the resulting publicity. Careers had been rightfully ruined, and there had even been the occasional criminal prosecution stemming from disclosures. However many more whistle-blowers had been jailed, persecuted, assassinated and in some cases even tortured to death in retaliation by criminal organizations and governments!
Story after story, and scandal on scandals had rocked the media through all sectors one by one. The Truth and Openness organizations, in spite of their su
ccesses, had few friends left, and fewer champions still standing. The sheer volume of the information, and the complexity of many of the accusations, had numbed much of the general public to the point of apathy. The gray line of information theft and misappropriation of private documents had been manipulated by the guilty parties with unlimited resources. Criminal charges had been laid on many Truth Movement members, as they were labeled terrorists, spies or anarchists by the state.