View PowerPoint Presentation
Wisdom [Meta-Knowledge] through AGI / Neural Macrosensing
“I believe that the creation of greater than human intelligence will occur during the next thirty years,” claims Vernor Vinge (1993: 1), mathematician, computer scientist and Hugo award-winning novelist. But what is the point of having more neural transmitters firing off connections if those connections do not promote wisdom?
Profound understanding, good judgment and deep insight, the elements of wisdom, are thought to be humanity's noblest goals. Western film icon and humorist Will Rogers wisely claimed that “Good judgment comes from experience, and a lot of that comes from bad judgment.” This insight gives credence to Buddha’s rumination that “We are what we think. All that we are arises with our thoughts. With our thoughts, we make the world.” (Brossman: 31)
Throughout my adult years, I have thought that the promising goal in life is to achieve wisdom, and that no matter how long or trying the pursuit, wisdom would be the accumulative reward. This may be true, but it is highly likely that there is a more direct route.
“Wherever it exists, wisdom shows itself as a perception of the relativity and relationships among things. It is an awareness of wholeness that does not lose sight of particularity or concreteness, or of the intricacies of interrelationships.” (Meeker 1981) Wisdom is a kind of meta-knowledge that helps us make better sense by putting our facts into a variety of contexts for viewing from a variety of perspectives—self-knowledge, laws-of-nature, and complex systems. In today's era of blending technology with human biology, scientists are speculating on the potential of biotechnologies for triggering these perspectives as we continue to try to make sense of the world and our relationship to it.
2. MAKING SENSE OF IT ALL
Is it possible to partner two distinct technologies for assisting our brain in acquiring wisdom? The maxim “two heads are better than one” takes on new meaning when the two are non-biological, engineered agents pooling their talents. Merging speculative technologies warrants careful consideration. Putting that important line of reasoning aside, but not forgetting it entirely, exploratory engineering is bringing about methods for assisting the human brain in higher-order cognitive processes. One approach is to couple two emerging, speculative technologies that have this potential, separately or together. These two technologies are artificial general intelligence (AGI) and nanotechnological macrosensing.
2.1. artificial general intelligence (“AGI”)
Artificial general intelligence echoes artificial intelligence's original aim of building machines with human-level and even superhuman intelligence. Unlike AI, which has become narrowly focused and task-oriented, (Voss 2006) AGI is based on the process of learning in order to gain knowledge, rather than being programmed with knowledge de facto. “In addition to optimizing the system for skills and knowledge acquisition and integration, [AGI] can also be designed to foster self-improvement, or intelligence bootstrapping. The most successful designs are likely to be Seed AIs, ‘designed for self-modification, self-understanding, and recursive self-improvement.’ (Yudkowsky 2000). Naturally, the system can also have full access to all of the Web’s facilities, plus substantial (distributed) human learning assistance.” (Voss 2001)
Engineers at Adaptive AI Inc. are building AGI that will act as intelligent assistants in various fields, including psychologically-coupled personal ‘oracles’. These systems will develop an intimate symbiotic relationship with individual humans—over time, becoming trusted advisors. The envisioned silicon oracle would assist its human counterpart on personal, intimate levels by interacting with our thoughts. According to inventor and scientist Peter Voss, "Once we have human level AGI, we will essentially possess our own personal AGI to integrate with us and advise us." Voss asserts that our new AGI silicon partner would develop rationality, wisdom, and knowledge through a relatively loose integration with our brain. This non-invasive approach to augmenting the brain would at first appear as mundane as an external, non-obtrusive black box, (which could give new meaning to “in the box thinking”); and later possibly as streamlined as light-activated ion channels for remote control of neural activity. (Richard H. Kramer) The oracle silicon partner would also be a new, emotional part of ourselves to bounce ideas off of; similar, but far more advanced than the Remembrance Agent (RA) computer, designed by Bradley Rhodes at MIT Media Lab, that watches over our shoulder and suggests information relevant to what we are reading or writing.
Voss believes that the AGI oracle will no doubt affect our thinking for the better. The AGI oracle would learn and/or be taught, not specifically programmed, and unique to each person; like an illusionary angel sitting on our shoulder, giving good advice. The oracle would be more like a complementary insider who listens to our emotional responses to the world around us and integrates with our cognition and be told when and how to interact with the individual. It might be proactive in giving advice, or it might reside unnoticed until petitioned to perform a task and suspended until its counsel or guidance is needed. In fact, it might be similar to Freud’s structural theory in which the oracle, as a superego, would be in touch with the elements of consciousness and work to balance their integration.
2.2 neural macrosensing
The second technology is "neural macrosensing." Macrosensing, a concept created by scientist and nanotechnology expert Robert A. Freitas Jr., relates to medical in vivo nanorobotics for the detection of somatic states as well as extrasomatic states, such as sensory data originating outside the body. In theory, it is the "ability to detect individual neural cell electrical discharges non-invasively … [and] offers the possibility of indirect ‘neural macrosensing’ of complex environmental stimuli by eavesdropping on the body's own regular sensory signal traffic." (Freitas 1999) Neural macrosensing would be comprised of millions of nanorobots in sensory locations designed to work with the entire body by listening to and recording sensory responses and transporting this information to the brain’s cognition center. Your brain would be in constant contact with your senses, allowing for super-senses or a more in tune ability to sense yourselves and your environment. (Freitas 1999) This means that nanorobots would listen to, or eavesdrop on, the body's sensory organs. The nanorobots would detect individual neural cell electrical discharges non-invasively and recognize and identify specific desired target nerve cells and pool data gathered independently by spatially separated nanodevices in real time.
For example, facultative (granted permission) mobile nanodevices may swim into the spiral artery of the ear and down through its bifurcations to reach the cochlear canal, then position themselves as neural monitors in the vicinity of the spiral nerve fibers and the nerves entering the epithelium (thin layer of cells) of the organ of auditory nerves within the spiral ganglion. These monitors can detect, record, or rebroadcast to other nanodevices in the communications network all the auditory neural traffic perceived by the human ear. … Since properly configured monitors can also modulate or stimulate nerve impulses, these devices may add audible signals to the audio traffic, thus may be employed as hearing aids (using feedback loops), real-time language translation mechanisms, continuous vocalization/audition recorders, voice-stress analyzers, or nano device-user communications links. (Freitas 1999)
“The most complex and difficult challenge in neural macrosensing will be optic nerve taps. The retina is thoroughly vascularized, permitting ready access to both photoreceptor (rod, cone, bipolar and ganglion) and integrator (horizontal, amacrine, and centrifugal bipolar) neurons.… Developing algorithms capable of interpreting raw optical nerve traffic, say, to recognize a specific human face or a specific scene in the vision field, would prove a significant research challenge.” (Freitas 1999)
2.3 AGI / neural macrosensing together
If AGI could be developed to assist the brain in decision making, it would have to be exceedingly knowledgeable about the body’s physiology, and most significantly its senses. The AGI oracle would need a full understanding of feelings, emotions and responses in order to objectively counsel its human counterpart. I propose that neural macrosensing nanorobots could provide the sensory data that the oracle needs. Since neural macrosensing is programmed to listen to the senses and direct data impulses to the thinking center of the brain, and AGI is a learning intelligence designed adapt to and assist the brain; it is possible that the two distinct technologies could work cohesively together. The following is a basic example:
Example 1: Jane is confronted with a person B who seems suspect. Because Jane is insecure, she wants everyone to like her and ignores the warning signals that her senses are picking up from her body’s reactions to stimuli. Her gut, or intuition, is signaling her brain that something is wrong, but she is not paying attention and shrugs it off by rationalizing the sensory input. There is a disconnect between the impulses her senses are sending to her brain and how her brain is translating the impulses into logical messages to her mind. If Jane were to have an AGI assist oracle, a different scenario could be like this:
Example 1a: Jane is confronted with a person B who seems suspect. The macrosensing nanorobots are working full time while Jane’s mind is busy covering up her suppressed fears. The body’s macrosensing nanorobots send the signals from varied locations throughout Jane’s body to the neural macrosensing nanorobots that collect the data from Jane’s ears, eyes, nose and vocal chords and then transmit the date to the AGI oracle. The oracle then taps into Jane’s cognitive properties with an assessment of the environment, and what visual and auditory nuances to pay attention to. The oracle might say something like this: “Jane, notice he repeated that word five times within 32 seconds. Warning: urine resides on his hands. Observe that he smells like cheap 1982 Merlot…”
2.4 AI / biological senses < AGI / neural macrosensing
Both artificial general intelligence and neural macrosensing are based on the inventions of exploratory engineering, similar to the innovation and futuristic inventions Leonardo da Vinci sketched out long before they could be realized. However, the theoretical coupling of AGI and neural macrosensing is based on current tangible advances in their respective fields. The motivation behind the advances is due in large part to overcoming hurdles within each technology and the overwhelming concern that computing power is advancing rapidly and could overtake human intelligence.
Marvin Minsky, father of AI said, “We have got to get back to the deepest questions of AI and general intelligence and quit wasting time on little projects that don't contribute to the main goal.” (Stork 2000) Stephen Hawking claims that computer intelligence will surpass human intelligence within a few decades and urges us to "develop as quickly as possible technologies that make possible a direct connection between brain and computer, so that artificial brains contribute to human intelligence rather than opposing it." (Perera 2001)
In 2006, the Artificial General Intelligence Research Institute Workshop convened just north of Washington, DC in Bethesda, Maryland. Speakers at the workshop are actively involved in designing and/or implementing AGI systems and offer relevant insight into the theory and practice of AGI. The momentum in this field, however slowly developing could generate a facelift on the narrowing field of AI. According to Ben Goertzel of Novamente, “There are probably a few dozen teams (some of them teams of one) in the world making deeply-serious efforts at achieving AGI.”
Interacting and learning through computer interfaces is different than face-to-face learning and has the potential for actually bringing about understanding and wisdom. AGI engineers hope that imagination and inspiration can be programmed into superintelligence, like Hal in 2001 Space Odyssey, along with the ability to form billions of calculations second by second as with “Deep Blue”. (Dreyfus 1993: 68, 103, 122). However, combining abilities of AGI and AI could bring this about.
2.4 Biological senses < macrosensing / neural macrosensing
In 2005 the National Institutes of Health grated Forty-Two Million Dollars to set up four nanomedicine centers. In April 2006, the journal Nature Materials estimated that 130 nanotech-based drugs and delivery systems were being developed worldwide. Even though nanomedicine is speculative, nanotechnology has enough bling to carry it along past conservative turnkeys. But “the possibility of using nanorobots in medicine, advocates say, would totally change the world of medicine once it is realized.” (Wikipedia 2006)
In 2005, Sander Olson of NanoTech.biz conducted an interview with Freitas.
The key issue for enabling full-immersion reality is obtaining the necessary bandwidth inside the body, which should be available using the in vivo fiber network I first proposed in Nanomedicine, Vol. I (1999). Such a network can handle 1018 bits/sec of data traffic, capacious enough for real-time brain-state monitoring. The fiber network has a 30 cm3 volume and generates 4-6 watts waste heat, both small enough for safe installation in a 1400 cm3 25-watt human brain. Signals travel at most a few meters at nearly the speed of light, so transit time from signal origination at neuron sites inside the brain to the external computer system mediating the upload are ~0.00001 millisec which is considerably less than the minimum ~5 millisec neuron discharge cycle time. Neuron-monitoring chemical sensors located on average ~2 microns apart can capture relevant chemical events occurring within a ~5 millisec time window, since this is the approximate diffusion time for, say, a small neuropeptide across a 2-micron distance. Thus human brain state monitoring can probably be ‘instantaneous’, at least on the timescale of human neural response, in the sense of ‘nothing of significance was missed.’
3. AGI ORACLE AND HUMAN COUNTERPART
But how would we secure a cooperative relationship between the oracle and its human counterpart? If someone of the stature of Stephen Hawkings states that “Humans must develop an interface that allows the human brain to be directly connected to a computer, so that the artificial brain contributes to human intelligence, rather than opposing it,” then it is a likely bet that we ought to pay attention. In fact, one way of doing this is through the suggested brain assist, such as an AGI oracle.
The most likely approach would be to build generic oracles with a large skill set and ability to bond quickly with their counterparts. This bonding would require more than technically-driven intellectual motivation; it would require strong sensory capabilities for, essentially, sniffing out the environment. And this is where neural macrosensing comes in; to "allow us to become exquisitely sensitive, like 'super-senses’ to fine details in our environment." (Freitas) Neural macrosensing could provide the needed sensorial feedback for both assisting the brain and developing elevated acuity.
Ben Goertzel is an optimist about the rate of progress toward AGI, and that AGI will most likely arise before neural macrosensing, and even be applied to drive neural macrosensing. Being a pragmatist, however, he is mindful of the possibility that neural macrosensing technology could advance faster than AGI, “in which case it will initially be driven by specialized ‘narrow AI’ (task oriented artificial intelligence rather than a learning general intelligence) applications rather than AGI.”
“Hybridizing this sort of narrow AI with neural
macrosensing technology could lead to amazing neuro feedback technology even
without AGI. Use of neuro feedback devices could allow us to control our
states of mind with far more fluidity and dexterity than is currently
conceivable, perhaps solving age-old problems such as the difficulty of
sustaining Zen-style ‘mindfulness’ while engaged in intensive and
challenging intellectual activity. [If we figured out the biological
indicators of the rare state of mind where both spiritual mindfulness and
intellectual mastery were being demonstrated, quite possibly neuro feedback
would allow us to learn to trigger that state via conscious effort.] And
the use of AGI for guiding neuro feedback, in place of relatively simple
machine learning methods, would make the possibilities even more awesome,
almost surely enabling states of consciousness that are currently
inaccessible (and literally unthinkable) to humans.” (Goertzel 2006)
Merging speculative technologies warrants careful consideration. Since a common building material for medical nanorobots is likely to be diamond or diamondoid substances, the first and most obvious question is whether diamondoid devices or their components are likely to be hazardous to the human body. (Freitas 2003) And achieving something as personal as wisdom may be viewed contrarily. Not everyone does or will consider the use of technology a dignified means for attaining wisdom. An article “The Age of Neuroelectronics” by Adam Keiper in The New Atlantis, a journal known as a ring of bioconservatives bent on opposing the cyberculture, Keiper writes “Without question, there are genuine human benefits to be gained if brain-machine technology advances in a sober, limited way. … Yet it is also possible to envisage a world where these new technologies are used for less noble purposes—from the next generation flight into alternative reality to the active manipulation and control of innocent subjects to the self-destructive pursuit of neurological perfection.” After a series of poking fun at decades of research on the brains, Keiper continues, “We are neither machines nor ghosts, but psychophysical unities—finite yet creative, embodied yet spiritual, cognitive yet not cognitive alone. No machine, however sophisticated, seems likely to duplicate or surpass that improbable mix of excellence, depravity, dignity, and uncertainty that is our human lot.” (Keiper: 4-41)
One thing is certain: change occurs, technology is man’s tool, and we want to be wise. We also like to make things pleasant, simple and relaxed. The finessing of wisdom makes the most difficult task seem effortless. The wise appear to glide through life, weightless. If technologies such as AGI and neural macrosensing could help our minds function more smoothly, with added ease and self-confidence, then there could be ample opportunities to observe the world around us. Seamless modification would make the pristine work of neural macrosensing seem so effortless that brain monitoring would be instantaneous while missing nothing. (Freitas, 1999) “We would be able to optimize our emotions with rational cognitive styles, awareness of our own mental activity, and careful evaluation of our worldview.” (More, 1999)
Optimizing emotions would be like adjusting a thermostat to turn up emotional heat or to turn it down “Since our immediate environment is rich in dangers and opportunities that range widely in importance, our brain needs something akin to a thermostat to determine when a specific challenge reaches the threshold of being sufficiently important to activate the several systems that focus attention and develop appropriate responses. Emotion, centered principally in a small set of sub cortical brain systems, is our biological thermostat, and so it's central to cognition.” (Sylwester 2001)
Cognitive scientist and AI pioneer John McCarthy, when questioned about what belief system his thermostat has, replied, "My thermostat has three beliefs. My thermostat believes that it’s too hot in here, it’s too cold in here and it’s just right in here." It sounds simple, and perhaps that is what wisdom ought to be.
Aristotle, "Metaphysica," translated by Ross, W.D., Richard McKeon (ed.), The Basic Works of Aristotle, Random House, New York, 1941.
Aristotle, Physics, 350 B.C. http://www.classics.mit.edu/Aristotle/physics.html
Blackmore, S., Consciousness: An Introduction, Oxford University Press, US, 2003.
Brossman, S., The Power of Oneness: Live the Life You Choose, Red Wheel, Newburyport, 2003.
Cohen, E., “Conservative Bioethics and the Search for Wisdom,” Hasting Center Report, Ethics and Public Policy Center, 2006. http://www.eppc.org/publications/pubID.2522/pub_detail.asp
Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999; http://www.nanomedicine.com/NMI.htm
Freitas R.A. Jr., Nanomedicine, Volume IIA: Biocompatibility, Landes Bioscience, Georgetown, TX, 2003; http://www.nanomedicine.com/NMIIA.htm
Freud, S., Beyond the Pleasure Principle, W.W. Norton & Company, New York, 1990.
Goertzel, B., Interview, 2006.
Keiper, A., “The Age of Neuroelectronics," The New Atlantis, 2006, pp. 4-41.
Kramer, R.H., “Light-activated ion channels for remote control of neuronal firing,” Nature Neuroscience 12:1381-1386, 2004.
Macdonald, C., “Playing the Wisdom Game,” Landscape, 1981, pp. 36-42.
MacDonald, C., Toward Wisdom: Finding Our Way to Inner Peace, Love and Happiness, Authors Choice Press, Lincoln, NE, 2001.
Meeker, J. W., Landscape, pp. 21-25, 1981.
More, M., “Mind Morph: Technologically Refined Emotions and Personality,” Extro 3 Conference, “The Future of the Body and Brain / Future Infrastructure, San Jose, CA, 1997.
Olson, S., “An Interview with Robert Freitas, Part II,” Nanoteh.biz, 2005. http://www.nanotech.biz/i.php?id=robertfreitas2
Nature Materials, “Nanomedicine: A matter of rhetoric?” Vol. 5, No. 4, p. 243m 2006. http://www.nature.com/materials/index.html
Stork, D., “Scientist on the Set: An Interview with Marvin Minsky,” Stork, D.G., (ed.), MIT Press, Cambridge, MA, 1996.
Perera, R., “Hawking: Re-engineer humans or risk machine rule,” CNN, Sci-Tech, 2001. http://archives.cnn.com/2001/TECH/ptech/09/05/humans.machine.idg/index.html
Rogers, W.P.A. http://www.cmgworldwide.com/historic/rogers/quotes5.htm
Sylwester, R., “Unconscious Emotions, Conscious Feelings, and Curricular Challenges,” 2001.
Voss, P., Interview, 2006.
Voss, P., “Essentials of General Intelligence: The direct path to AGI,” Artificial General Intelligence, Goertzel, Pennachin, Cassio (Eds.) 2006.
Vinge, V., “The Coming Technological Singularity: How to Survive in the Post-Human Era,” VISION-21 Symposium, NASA, 1993.
Yudkowsky, E., “What is Seed AI,” 2000. http://www.singinst.org/seedAI/seedAI.html