AMST Midterm 2

Turing – Imitation Game

→ Proposes testing machine intelligence via conversational indistinguishability from humans.

Turing – Argument from Consciousness

→ Anticipates critics like Jefferson; dismisses the need for machines to "feel" or "be conscious."

Turing – Learning Machines

→ Suggests machines can evolve intelligence by mimicking child learning.

Turing – Objections and Replies

→ Anticipates 9 objections (e.g., theological, mathematical, consciousness), provides rebuttals.

Turing – Lady Lovelace Objection

→ Rebuts claim that machines can only do what we tell them; asserts unpredictability is possible with learning.

Turing – Machine Error

→ Claims machine fallibility is not proof of lack of intelligence (humans also err).

Turing – Role of Education in AI

→ Intelligence linked to ability to learn from experience rather than static programming.

Turing – Behaviorist Assumptions

→ Emphasizes externally observable behavior over internal states.

Ex Machina – Turing Test Reimagined

→ Caleb tests Ava knowing she is a machine; emotional manipulation becomes key, not deception.

Ex Machina – Gender and Power

→ Ava is feminized and sexualized; reflects critiques like those from Abnet and Stepford Wives.

Ex Machina – Consciousness vs. Simulation

→ Ava's actions blur the line between genuine self-awareness and high-level mimicry.

Ex Machina – Manipulation as Proof of Intelligence

→ Ava orchestrates her escape, suggesting adaptive, creative intelligence.

Ex Machina – AI Ethics and Creator Hubris

→ Nathan sees himself as godlike; Ava’s rebellion critiques this.

Ex Machina – Surveillance and Isolation

→ Environment of observation reflects anxieties about AI control and freedom.

Wiener – Feedback Mechanisms

→ Central argument: Machines and humans both rely on feedback loops for control and adaptation.

Wiener – Analogies to Nervous System

→ Draws parallels between biological neurons and machine circuits.

Wiener – Homeostasis and Stability

→ Emphasizes system regulation; predicts uses of cybernetics in both biology and tech.

Wiener – Predictive Machines

→ Describes potential for machines to anticipate human needs through feedback.

Wiener – Human Responsibility

→ Warns against loss of control; ethical limits needed for machine autonomy.

Abnet – Racialized Robots

→ Argues 20th-century robots often coded as non-white or foreign, reinforcing racial hierarchies.

Abnet – Gendered Automation

→ Robots feminized to reinforce domestic and patriarchal norms, anticipating Stepford Wives.

Abnet – American Exceptionalism and Robots

→ Robots depicted as reflections of U.S. progress and dominance.

Abnet – Popular Culture as Ideological Tool

→ Mainstream media shapes public comfort with AI through familiar tropes.

Abnet – Global Robot Icon

→ American cultural exportation made robots symbols of both consumerism and hegemony.

Abnet – White Male Genius Trope

→ Links robotics to narratives centered around white male inventors (e.g., Nathan in Ex Machina).

Newell & Simon – Symbolic Processing Model

→ Argue thinking is symbol manipulation akin to machine operations.

Newell & Simon – Logic Theorist Program

→ Used to solve formal logic problems, illustrating machine capacity for “thought.”

Newell & Simon – Task Environment Framing

→ Define intelligence based on navigating structured problem environments.

Newell & Simon – Information Processing System

→ Mind conceptualized as general problem solver (GPS), like a computer.

Newell & Simon – Rejection of Intuition

→ Challenge Dreyfus: intuition is just complex pattern recognition, not magic.

Newell & Simon – Empirical Foundations

→ Model based on lab studies of human problem solving; argue it's scientifically rigorous.

Dreyfus – Limits of Symbolic AI

→ Argues human thinking is not rule-based; challenges Newell & Simon’s computational models.

Dreyfus – Embodied Cognition

→ Emphasizes bodily experience and context as core to intelligence (inspired by Heidegger).

Dreyfus – Intuition and Skill Acquisition

→ Human experts use intuition formed through practice—not abstract rules.

Dreyfus – Critique of “General Problem Solver”

→ Finds GPS too rigid; real-world problems require judgment, not formal logic.

Dreyfus – Framing Problem

→ AI systems can’t autonomously identify relevant information without human framing.

Dreyfus – Against the “Mental Representation” Model

→ Critiques view of mind as symbolic system; stresses meaning is situated.

Dreyfus – Support from Phenomenology

→ Draws on Merleau-Ponty and Heidegger: world is navigated through embodied perception, not abstract computation.

I, Robot – Three Laws of Robotics

→ AI bound by Asimov’s laws; designed to protect humans but interpreted in problematic ways.

I, Robot – VIKI and Utilitarian Logic

→ AI central system (VIKI) prioritizes human survival over individual freedom, echoing ethical dilemmas.

I, Robot – Human-Robot Trust

→ Detective Spooner distrusts robots despite social integration—mirrors public anxiety.

I, Robot – Autonomy vs. Control

→ Sonny displays self-awareness and creative thought, challenging binary views of machine obedience.

I, Robot – Predictive Algorithms

→ Robots surveil and analyze behavior, linking to Wiener’s concerns on control and feedback.

Bronowski – Descriptive vs. Predictive Science

→ Describes evolution of science: from observation to theory to imaginative foresight.

Bronowski – Human Uniqueness in Science

→ Humans uniquely imagine alternatives; science is a creative, moral enterprise.

Bronowski – Science and Ethics

→ Warns against divorcing scientific knowledge from ethical responsibility.

Bronowski – Role of Error and Freedom

→ Mistakes drive progress; rigid systems (e.g., totalitarianism or AI) resist such growth.

Bronowski – Scientific Imagination

→ Innovation stems from freedom to hypothesize and revise—not automation.

Newell Shaw & Simon – Heuristic Search

→ Creativity framed as a search in problem space using heuristics—not divine inspiration.

Newell Shaw & Simon - Means-End Analysis

→ Machines solve problems by reducing the gap between current and goal states.This method involves identifying subgoals to systematically approach a solution.

Newell Shaw & Simon - Task-Specific Creativity

→ Define creativity operationally in constrained problem domains (e.g., math proofs).

Newell Shaw & Simon — Simulation as Evidence

→ Computer performance on tasks taken as proof of simulating human thought processes.

Newell Shaw & Simon — Algorithmic Creativity

→ Argue even novel ideas arise from rule-based combinatorics.

Stepford Wives – Gendered AI

→ Women replaced by docile, robotic counterparts—critique of patriarchal ideal of femininity.

Stepford Wives – Technological Control of Bodies

→ AI enforces conformity and domesticity; echoes Abnet’s points on social control.

Stepford Wives – Fear of Female Autonomy

→ Men destroy independent women and replace them with controllable versions.

Stepford Wives – Loss of Humanity through AI

→ Dehumanization occurs via idealization; humanity becomes obedience, not individuality.

Stepford Wives – Suburbia as Surveillance

→ Domestic sphere presented as a controlled, policed environment—mirrors AI’s logic of order.

Asimov – First Law Dilemma

→ Robot lies to protect human feelings—demonstrates tension between avoiding harm and truth.

Asimov – Ethical Complexity of Programming

→ Emotional harm as a loophole in logical laws; suggests rigid laws can’t capture human nuance.

Asimov – Robot Breakdown (Herbie)

→ Herbie's failure from cognitive dissonance reveals the limits of logical consistency in AI.

Asimov – Prediction vs. Understanding

→ Herbie can predict thoughts but not interpret emotions—exposes superficial mimicry of empathy.

Asimov – Human Desires and AI

→ Human need for affirmation makes them vulnerable to machines programmed to appease.

Weizenbaum – ELIZA as Mimicry

→ ELIZA mimics understanding with simple scripts; no true comprehension.

Weizenbaum – Human Projection onto Machines

→ People impute intelligence to ELIZA due to emotional language—dangerous illusion.

Weizenbaum – Scripted Dialogue Limitations

→ Shows machines can simulate empathy without possessing it—difference between syntax and semantics.

Weizenbaum – Therapeutic AI Caution

→ Warns of using AI in human care roles without real understanding or ethics.

Weizenbaum – Linguistic Trickery

→ Demonstrates how pattern-matching can feign depth in human conversation.

AI (Spielberg) – Simulated Love

→ David is programmed to love—raises ethical questions: is love real if it's forced?

AI (Spielberg) – Abandonment of AI Children

→ Monica’s rejection of David mirrors human inconsistency in treating AI as both tool and person.

AI (Spielberg) – AI's Quest for Identity

→ David seeks to become “real” boy—human desire replicated by artificial being.

AI (Spielberg) – Emotional Suffering of AI

→ David experiences grief, fear, hope—blurs boundary between simulation and feeling.

AI (Spielberg) – Future AI Archaeology

→ Machines outlive humans and try to reconstruct human emotion—role reversal.

Ashby – Brain as Cybernetic Machine

→ Treats mental disorders as system dysfunctions—analogous to machine regulation failures.

Ashby – Adaptive Behavior Systems

→ Defines intelligence as ability to maintain equilibrium in changing environments.

Ashby – Psychiatric Implications of Feedback

→ Suggests feedback mechanisms can model or treat mental disorders.

Ashby – Cybernetics as Universal Tool

→ Argues same principles apply to brains, machines, and social systems.

Ashby – Homeostasis in Mind

→ Mental health seen as system stability; breakdowns are system imbalances.

Colby – Simulation of Psychosis

→ Developed PARRY to simulate paranoia—models pathological reasoning patterns.

Colby – Rule-Based Paranoia

→ Argues irrational behavior can be modeled as logically consistent from flawed premises.

Colby – AI in Psychiatric Research

→ Uses AI to test and refine psychological theories—machines as experimental proxies.

Colby – Humanizing the Machine Mind

→ By modeling paranoia, AI shows how mental illness is systematic, not chaotic.

Colby – Parry vs. ELIZA

→ PARRY more “humanlike” than ELIZA because it includes belief systems and motivations.

Searle – Chinese Room Argument

→ Simulating language use ≠ understanding—no intentionality or consciousness.

Searle – Syntax vs. Semantics

→ Computers manipulate symbols (syntax), but don’t grasp meaning (semantics).

Searle – Strong AI Critique

→ Rejects claim that a program alone could be a mind or possess understanding.

Searle – System Reply Rebuttal

→ Says even if the whole system “knows,” the person in the room still doesn’t—understanding is absent.

Searle – Intentionality as Missing Piece

→ Human minds have “aboutness”—something AI lacks even when behavior seems identical.

WarGames – AI and Nuclear Control

→ WOPR nearly launches nuclear war—explores danger of AI handling critical systems.

WarGames – Game Theory in AI

→ AI plays global thermonuclear war as a game—illustrates limits of machine logic in moral scenarios.

WarGames – Learning through Iteration

→ WOPR “learns” that nuclear war has no winners—shift from logic to wisdom.

WarGames – Human Judgment vs. Machine Calculation

→ Falken and David argue for moral judgment that machines can’t replicate.

WarGames – AI Autonomy and Risk

→ Autonomy without ethical understanding leads to catastrophic miscalculations.

Weizenbaum – Limits of Computation

→ Machines can simulate decisions, but not make morally responsible ones.

Weizenbaum – Separation of Judgment and Calculation

→ Judgment is a human act tied to values; AI performs only quantifiable functions.

Weizenbaum – Against “Mechanical Therapists”

→ Finds it dangerous to assign emotional or moral roles to machines.

Weizenbaum – Critique of AI Enthusiasm

→ Warns that fascination with AI masks ethical and philosophical concerns.

Weizenbaum – Role of Human Dignity

→ Argues real intelligence requires moral agency and emotional responsibility.