The Hard Debate, Senior Year
Standard quoted exactly
Debate perspectives on differences between human and artificial intelligence and their implications for consciousness, creativity, ethics, and human responsibility.
Student-friendly learning targets
- I can debate differences between human and artificial intelligence using evidence from prior units, not slogans.
- I can address all four implications named in the standard: consciousness, creativity, ethics, and human responsibility.
- I can steelman a view I do not hold and still keep humans accountable for design, use, and harm.
Essential questions
- What is actually different between human minds and current AI systems, and which differences matter for policy?
- If a system can produce a poem, a plan, or a kind sentence, what follows for consciousness, creativity, ethics, and responsibility — and what does not follow?
- Who remains responsible when a fluent system is in the loop?
Objectives
- Revisit 9-12.AIF.ECT.1 at greater depth: same prompt, higher evidence bar, senior-year sources.
- Present an evidence-backed claim on each of the four implications: consciousness, creativity, ethics, and human responsibility.
- Use at least one technical limit (pattern completion, lack of bodily sensing, hallucination, no account-holding) and one civic consequence.
- Steelmans the opposing view for two minutes before rebuttal.
- Refuse vibes: every major claim carries a dated source, a prior-lesson artifact, or a clearly labeled philosophical axiom.
Key vocabulary
- Consciousness
- There being something it is like to be a subject — experience, not only intelligent-looking behavior. Whether any current system has it is disputed; pretending the dispute is over is not evidence.
- Anthropomorphism
- Attributing human feelings, intentions, or inner life to a system because it talks like a person.
- Creativity
- The making of something new that is also meaningful in a human practice (art, science, craft). Novelty of output is not automatically creativity of a subject.
- Moral patient / moral agent
- A moral patient can be wronged; a moral agent can be held responsible. Debate which, if either, applies to AI, and why humans remain agents either way.
- Human responsibility
- The duty of people and institutions to set aims, check outputs, take blame, and repair harm. It does not migrate to a model because the model is fluent.
- Steelman
- The strongest fair version of an opposing view, stated so that a serious person who holds it would recognize it.
- Technical limit
- A documented constraint of current systems (no body, no persistent account-holding, training on human traces, hallucination) used as evidence rather than as insult.
Teacher background
Standard 9-12.AIF.ECT.1 asked ninth and tenth graders to debate human versus artificial intelligence and the implications for consciousness, creativity, ethics, and human responsibility. This is the same sentence at senior depth. The bar is evidence, not vibes. Students should walk in with artifacts from earlier AII lessons: accessibility failures, energy and water costs, unauthorized training, dual-use cybersecurity, three-test evaluations, career loops, and stage-by-stage bias. Those artifacts are exhibits. Consciousness: distinguish inner experience from skilled imitation of talk about experience; name the dispute; do not grade students for being dualists or functionalists, grade them for not smuggling a conclusion. Creativity: a model can produce a new combination; a human practice still decides whether it is a work, a tool output, or a derivative injury to a living artist (EP.4). Ethics: duties attach to agents who can answer. Human responsibility: even if someone argues a future system might be a moral patient, today’s deployments are designed, bought, and excused by people — school boards, vendors, nurses, engineers, students. Require a steelman. Ban diagnosis of classmates’ spirituality. Civic tone: this is a seminar, not a TED talk and not a panic session. Offline fallback: printed exhibit cards from prior units and timed debate blocks that need no network. FERPA: personal journals stay personal; cite class artifacts without naming a peer’s private beliefs.
Materials and prep
Materials
- Four exhibit folders, one per implication, stocked with dated cards from prior lessons (EP.2 door, EP.3 warehouse, EP.4 scrape, EP.5 dual use, IS.1 protocol, IS.2 loop, IS.3 stages).
- Debate roles: opening, steelman, rebuttal, civic implication, timekeeper.
- Evidence rule card: source, date, or labeled axiom. Vibes will be stopped.
- Timer and speaking-order poster.
- Offline fallback: paper exhibits only. No live chatbot asked Are you conscious? as a stunt.
Before class
- Tell students two days ahead to bring one prior-unit artifact. Have extras.
- Assign mixed groups so the four implications are all staffed.
- If 9–10 debate notes exist, return them so growth is visible.
- Block plan: full structured debate with judge sheets, then a 25-minute argumentative essay covering all four implications.
Instructional sequence
Vibes are not exhibits
5 min- Students write one sentence they have heard: AI is just a tool, AI is already alive, AI will take everything, only humans create. They mark it vibe or exhibit. If they cannot name a source, it is a vibe.
- Collect two vibes and two exhibits on the board. Leave the vibes up as caution tape.
- Read the standard. Point to the four nouns. All four will be hit or the debate is incomplete.
- Remind them this revisits AIF.ECT.1 on purpose.
Four implications, one evidence bar
10 min- Consciousness: define the dispute. Behavior can fool. Inner life is not proven by a first-person sentence a model emits. Students may argue either way if they carry a philosopher, a scientist, or a clearly labeled axiom — not a movie.
- Creativity: return to EP.4. Output can be novel; credit, intention, and practice still sit with humans unless someone argues a careful alternative and takes questions.
- Ethics: who can be wronged, who can owe a duty. A crashed model is not punished; a vendor, a board, or a user can be.
- Human responsibility: the through-line of the year. Designers, deployers, and users choose. The algorithm is not a legal person in this classroom’s working assumption unless a student argues otherwise with sources.
- Teach steelman with a 30-second model. Teach the technical-limit requirement.
Steelmans on consciousness
10 min- In pairs, Student A states a consciousness claim in one sentence with a source or axiom tag.
- Student B steelmans it until A nods. Then B offers one technical limit (for example, no interoception, no persistent body, training on human text).
- Swap. Teacher listens for mockery and stops it.
- Harvest one civic consequence: if we wrongly treat a system as a person, what duty to humans might we dodge? If we wrongly treat people like systems, what duty might we dodge?
Four-box brief before speaking
12 min- Each student completes a four-box brief: Consciousness / Creativity / Ethics / Responsibility, each with a claim, an exhibit, and a steelman seed.
- Missing a box means they are not ready to speak on the full standard.
- They star the one box they will open with in the hands-on debate.
- Teacher stamps briefs that cite prior units.
Real-world examples
- A student who treats a companion chatbot as a friend (AIF companion risk) versus a student who treats it as a drafting tool — implications for consciousness talk and for responsibility.
- An Idaho artist’s unpaid style in a generator: creativity as practice and market, not as a vibe about the muse.
- A hospital dual-use failure (EP.5) where blaming the model would dodge a human protocol.
- A school board asked whether a detector is fair (IS.3 stages) — ethics without metaphysics.
- Senior capstone work: who is the author of record if a model drafted a paragraph the student then rebuilt.
Hands-on activity
Four corners, timed
8 min- Room corners: Consciousness, Creativity, Ethics, Responsibility. Groups of four rotate or send a runner so each corner hears a 60-second evidence claim and a 30-second steelman.
- A scribe at each corner notes one technical limit and one civic consequence.
- If time is short, stay in home groups of four with one student owning each implication.
- Collect four-box briefs. Oral performance is formative; the brief is the gradeable artifact in a 50-minute period.
Discussion questions
- If a system says I feel afraid, what kind of evidence would you still need before talking about consciousness?
- Is a new image that no human composed pixel by pixel a creative work, a tool output, or something the law has not settled (link EP.4)?
- Can a model be a moral patient without being a moral agent? What would that change in a school policy?
- When people say the algorithm decided, which human choices are being hidden?
- What did you believe in the 9–10 version of this debate that you can no longer hold after EP and IS evidence?
Differentiation
Support
- Provide a four-box template with sentence frames and a list of allowed exhibits from prior units.
- Allow a written steelman instead of public speaking, still timed.
Challenge
- Require engagement with a named philosopher or scientist (dated, excerpted by the teacher) in the consciousness or ethics box.
- Write a dissent that still hits all four implications.
Multilingual learners
- Permit the steelman in the student’s strongest language with an English claim sentence, or vice versa.
- Pre-teach consciousness, agent, patient, and responsibility with non-idiomatic definitions.
IEP / 504
- Do not force unprepared public debate. The four-box brief meets the standard; speaking is an extension.
- Extra processing time on the steelman. Quiet copies of exhibit cards.
Assessment
Formative
- Vibe versus exhibit warmup.
- Paired steelman on consciousness.
Summative
- Four-box brief scored on all four implications, evidence bar, steelman seed, and a technical limit plus civic consequence.
- Block extension: judged debate plus a 500-word essay that a 9–10 student could not have written without the year’s exhibits.
Success criteria
- All four implications appear with a claim.
- At least one technical limit and one civic consequence are explicit.
- Steelmans rather than caricatures; no vibe-only boxes.
Responsible use, ethics, and privacy
Responsible use
Do not ask a live model whether it is conscious as a classroom stunt; that confuses the evidence bar. Do not pressure students to disclose religious views of the soul. Paper exhibits from prior lessons are enough. Offline is the default.
Ethics
The ethical implication is not only how we treat machines. It is how talk about machines becomes a way to dodge duties to workers, students, patients, and creators. Students may disagree about inner life. They may not disagree that today’s harm and credit still land on people unless they argue otherwise with care.
Privacy
Beliefs about the soul, mental health, and companion-app use can be sensitive. Students cite public class artifacts, not a neighbor’s journal. FERPA still covers recorded student work if you keep debate video; store it under district rules. No uploading of student briefs to a scoring bot.
Reflection
- Which implication was hardest to keep from collapsing into a movie plot?
- What exhibit from this year changed a 9–10 opinion?
- Where did you catch yourself anthropomorphizing, and what did you do next?
Homework
Revise the four-box brief. Add one quotation from a prior lesson packet (with lesson code) to the weakest box. Do not ask a chatbot to steelman for you; that would skip the thinking the standard names.
Closing
Read the four nouns again. Collect briefs. The last lesson in this subdomain asks the civic sequel: who decides how these systems are used, designed, deployed, and regulated — and who pays.
Extensions and cross-curricular links
Go further
- 90-minute block: full debate with judge sheets and a sustained essay (seminar and writing).
- Invite a philosophy or CS faculty guest from a nearby college; recorded backup.
- Return 9–10 debate notes and write a metacognitive postscript: what evidence changed.
- Publish (with student consent, no names required) a class list of remaining disagreements — intellectual honesty as a product.
- Philosophy / English seminar
- Personhood, authorship, and the difference between a claim and an axiom.
- Psychology
- Anthropomorphism, theory of mind, and why fluent talk triggers social brainware.
- Civics
- Responsibility in institutions when software is in the loop.
- Arts
- Creativity as a practice with audiences, not only as surprise in a file.