Showing posts with label computer intelligence. Show all posts
Showing posts with label computer intelligence. Show all posts

Wednesday, August 2, 2023

INTELLIGENCE AND EVOLUTION

Natural selection is the basis of evolution,  a basic tenet of biology.  Natural selection occurs by genetic selection in mating individuals. This theory includes the evolution of the adaptive capacity of "intelligence".  Most investigations of "intelligence" explore how this feature of brain operations developed in the course of primate evolution.   The selection of genomes as "more fit" is not absolute but related to a specific environmental parameters,  and variations in genome are selected to optimize this "fit".  

With the development of more complex behaviors and language, an associated "culture" accompanies humans from generation to generation,  and with the development of writing and more advanced methods of information storage,  it is possible to store and transmit this "culture" for the forsee-able future. "Culture" is not transmitted in the genome,  and the Lamarkian theory that behavior contributes to natural selection is rejected in evolutionary theory.  This leaves the question of whether (and how) "cultural knowledge" contributes to adaptation, and evolution.   Several theories of "cultural evolution" involving "memes" have been proposed.  It is not clear if changes in memes are efficiently selected for adaptive value.  Questions might be raised about whether one economic system is more adaptive because it is more "productive".  Does maximizing "productivity" represent the highest level of adaptive capacity,  or are other variables involved which must also be included?  Is "religion" an adaptive cognitive survival skill?  Given the extensive mass killings, historically associated with religions,  it is unclear how this promotes adaptive survival.  Is the dominance of one religion explained by "survival of the fittest"?

Science would seem to be the area of culture most clearly associated with adaptive value,  and many scientific achievements have contributed to prolonging human life.  But the relationship of science to evolution is growing more complex, and several recent achievements raise troubling questions about how evolution is influenced by scientific achievements.  These include: 1) nuclear weapons and radiation, 2) gene modifications and CRISPR, 3) cloning of organisms,  and 4) AI repurposing of cognitive activity.

Discoveries in the nuclear structure of matter are a prominent achievement of 20th century physics.  Separate from this investigation of basic knowledge,  nuclear fission and fusion reactions can release large amounts of energy, and were studied for use in creating weapons.  These weapons exist, have been used, and are relatively widely distributed in the developed countries.  Their use threatens the genetic integrity of those exposed,  and these genetic effects including later development of various cancers is well documented.  The risk of the widespread uncontrolled use of these weapons is an evolutionary threat, so it makes  sense to manage the control of their use.  Whether this is the current situation is unclear, and no formal system for preventing their unrestricted use exists,  and evidence for individual leaders with a poor reputation for appreciation of consequences who have threatened the use of these weapons raises serious questions.

There is a long history of genetic modification of plants and animals by selective breeding to improve the value of plants for human consumption, and animals as domestic assistance or for food.   This distorts the evolutionary viability of these organisms,  and usually makes the animals dependent on human management for survival.  More subtle genetic modifications are occurring as animals adapt to human urban environments.  Two more dramatic interventions are cloning and CRISPR.  In cloning,  animals (or humans) are reproduced from selected genetic material.  This is the extreme of "eugenics", not breeding for ideal offspring,  but cloning yourself to be the offspring!  The assumption that the clone would be identical to the original person takes into account only the genetic component,  and the role of environmental programming cannot be predicted in the same way and is unlikely to reproduce the same personal configuration.  The danger of cloning is its disregard for recombinant genetic variation.  Cloning an individual is the ultimate narcissistic statement that indicates the person is ideal, and no modifications in their genome are needed.  Cloning of animals (or plants) sometimes occurs in Nature,  but for more complex animals it does not,  as adaptive selection is prevented.  Most troublesome is CRISPR.  The ability to modify plants and animals by introducing genetic material by non-sexual selection methods ENTIRELY BYPASSES EVOLUTION in favor of "scientific" presumptions of better adaptive value.  Its use has been proposed for correcting single locus genome abnormalities in human illnesses,  where somatic cells can be modified,  and also proposed for as yet unachievable interventions in embryos.  The medical application seems humanitarian and does not automatically alter evolution.  But the urge to find genetic material that enhances human capacities,  and insert this to produce "ideal" embryos is a lurking danger.  Here again, bypassing the evolutionary process produces unknown non-evolutionary consequences.

AI is the ability to combine and integrate human cultural production to produce new cultural "products".  Since all human culture bypasses evolutionary selection,  it is not clear how this particular modification poses risk other than more direct cultural interventions.  In a recent opinion, Karp  has compared the development of nuclear technology to AI. (https://www.nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html)  To understand the danger of AI,  one must understand that human cultural products are created by individuals.  Each person receives training from others about basic cultural memes and how to use them to participate in the social system in which he/she is embedded.  The evolutionary impact on culture depends on how cultural elements appeal to and manage the  motivations of individuals.  When this impact is mal-adaptive, the long term result is a contraction, or elimination of the culture.  Economic rewards,  sexual selection, and social support are all linked to appropriate performance of cultural behavior.  (The failure to consistently conform is usually considered "mental illness".) In more complex societies,  bureaucratic entities,  governments, corporations, etc, are given legal existence to have power to enforce and reward cultural performances.  These artificial entities may demand cultural behaviors that are mal-adaptive (that interfere with evolution).  When evolution is effective,  mal-adaptive behavior of these "artificial organisms" results is eventually selected out.   What makes robotics/AI so threatening is the possibility of creating entities that perform actions with a significant level of human intellectual capacity,  but are not regulated by human motivations. What then are they responsive to?  The current fear is that they are responsive to corporations that are not empathic to the needs of the larger society.  This is the (alleged) effect of social media corporations and other technological developments that have emerged in recent years.  But it is unclear if they are disregarding the larger society, or responding to conflicting influences  with different political interests.  (An intelligent governmental response in this country appears to be impossible due to the current technological ignorance of the political leadership.)  The danger of AI is that human intellectual capacities can be mimicked by computational devices that humans will be unable to discriminate from interacting with another human, and so will be unaware of being manipulated by some entity.  The evolutionary extreme would be humans  manipulated to "fall in love with" pornographic AI productions which distract them from human reproduction and materially alter the ability of evolution to influence natural selection.  (How much this already occurs in dating sites is unknown.)

So long as intelligence is closely linked with human motivations in individuals susceptible to social and sexual selection,   it is indirectly influenced by evolution,  including its "cultural" component.  Any intellectual activity that bypasses the evolutionary selection process threatens to alter evolutionary processes with unknown consequences!

Sunday, March 12, 2023

The storm over ChatGPT

The Wall St journal recently published an article "ChatGPT Heralds an Intellectual Revolution" which is possibly prophetic, but more likely promotional hyperbole.  The company has recently released its product to general use, and two of the three authors are highly invested in tech companies (but do not disclose whether they are invested in this one or not). Other language AI systems are in the works in other tech companies,  and this may be an effort to front-load this version for investment..   The technology combines strings of verbal expressions into sentences, and paragraphs, which are perceived as responsive to questions presented to the system.  A summary of features by Business Insider includes many pitfalls (https://www.businessinsider.com/everything-you-need-to-know-about-chat-gpt-2023-1) (and the Microsoft version may involve the authors financially).  ChatGPT can also be used as a text generator for school essays, applications, etc.  No one has yet proposed that it be used to generate legal documents, nor to write "scientific papers" with a data source.  To some, the ability to generate "human sounding" text that is related to an inquiry seems a high level of "human intelligence".  The authors of the WSJ piece compare its importance to the Gutenberg printing technology and the emergence of the enlightenment.  Similar pronouncements have accompanied other tech advances like the internet,  which in retrospect provides both opportunities and challenges to its effective use. (The same might be said of any technology including the printing press.)  

Chomsky has written a blistering critique of the technology claiming that it simplifies more complex human intellectual generative processes: (https://www.nytimes.com/2023/03/08/opinion/noam-chomsky-chatgpt-ai.html )  OpenAI’s ChatGPT, Google’s Bard and Microsoft’s Sydney are marvels of machine learning. Roughly speaking, they take huge amounts of data, search for patterns in it and become increasingly proficient at generating statistically probable outputs — such as seemingly humanlike language and thought. These programs have been hailed as the first glimmers on the horizon of artificial general intelligence — that long-prophesied moment when mechanical minds surpass human brains not only quantitatively in terms of processing speed and memory size but also qualitatively in terms of intellectual insight, artistic creativity and every other distinctively human faculty.

Chomsky emphasizes the more sophisticated search and organization features of the human process,  which depend less on brute statistical operations.   Also to the point is the question posed by Wladawsky-Berger "Are GPT3 and Chat GPT truly intelligent?" (https://medium.com/mit-initiative-on-the-digital-economy/are-gpt-3-and-chatgpt-truly-intelligent-38475c0c1692) In their paper, Bender and Koller explain why LLMs like GPT-3 are likely to become innovative language tools — kind of like highly advanced spell checkers or word processors — while dismissing claims that they have the ability to reason and understand the meaning of the language they’re generating. Their explanations are couched in linguistic concepts which they carefully define: form, communicative intent, meaning, and understanding.

Viewed from this perspective the WSJ article does seem a promo for investment.  But the interesting question of "true intelligence" is raised by these systems.   ChatGPT demonstrates that a specific capacity of human function, the ability to link together verbal information in an apparently meaningful (to humans) pattern can be simulated by computers. What more do humans add and how do we do it?  The obvious answer is that we add data from our bodies, somatic patterns,  hormones, "emotional reactions", in summary a range of responses that are not purely verbal and cannot always be easily translated into verbal form.  (That this might not immediately occur to computer scientists is not surprising.)  We also add the data of aggregate sensory experiences, the combination of signals entering the body through sense organs as we move around the world.  None of this is directly available to ChatBots and must be translated into verbal patterns that approximate human experience.  The result is that the output of the ChatBot is never time stamped.  It is not about the current experiential space though it may seem as if it is.  There are other differences as well, so it is surprising that some humans are impressed by the limited accomplishments of this interactive verbal process.   Do they suppose that ChatBots can replace meaningful human exchange?  Can ChatBots be used to engage patients in psychotherapy?  negotiate contracts? interrogate possible criminals?  There is no evidence that they have the ability to perform any of these tasks.  Their major contribution is showing that verbal exchange based on combining random verbal information is a weak simulation of a facet of human intelligence.

The more extensive question of computation and intelligence was raised by Turing (A summary in https://medium.com/@jetnew/a-summary-of-alan-m-turings-computing-machinery-and-intelligence-fd714d187c0b.  

Full text PDF: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjVgo_bo9f9AhXbHkQIHcSBBjUQFnoECA4QAQ&url=https%3A%2F%2Fredirect.cs.umbc.edu%2Fcourses%2F471%2Fpapers%2Fturing.pdf&usg=AOvVaw1WyXuTomtTJiRGVWr5k5R3).   

Turing explored the range of issues as an interrogation by human of either a computer simulation or another human.  The analysis illustrates the fundamental issue of human intelligence: it is about the ability to interact with another intelligence in real time in a manner that can effectively influence the other, over a range of situations. (Computers can effectively play chess, go, and Jeopardy against humans, but this range of situations is very limited.)  Current technologies are far from this capacity,  but apparently can be used effectively to sell themselves in online opinion segments!