The idea map

Last updated 2026-08-10 · build 20260810T1806Z-afe88a6c · Markdown version · corpus index

The ideas ring carries this initiative’s arguments in full — each compressed from the primary sources, with dated lineage, verbatim passages, and honest reception. This page is the map behind it: the whole territory those arguments are drawn from, one line per thread, grouped by subject. Treat it as a broad map of the territory the Future of Being Human initiative covers, works within, and contributes to. As with all maps, there is more detail presented in this corpus in some areas than others.

The grouping is part of the work the initiative does and what it represents. We hold that navigating advanced technology transitions is combinatorial work, and that the connections across these groups are where the contribution lies rather than in any single vertical (Philosophy). Read down a group and you see a line of thought; read across them and you see the architecture the ring’s cross-links trace. The territory is shared: these threads are the work of both principals, Andrew Maynard and Sean Leahy, and of the conversations between them — a great deal of the thinking here is made on Modem Futura rather than reported there.

Threads carrying a link have a full page in the ring. The rest live where they were made — in the essays, books, papers, podcast episodes, and reports the ring draws on — and are named here so the shape of the work is visible rather than asserted. This is a map of ground covered and active areas of thinking and practice. It is not a schedule of work. As the work of the initiative evolves, some threads will become pages, some will stay as they are, and some will turn out to be facets of others.

Core thesis — the future of being human

Why the question matters now, and what is actually changing about the human condition.

• The inflection point — complex systems undergo sudden, hard-to-reverse transitions, so “it always worked out before” is structurally invalid reasoning about AI.
• Base codes and transcoding — mastery over the digital, biological, and material codes of reality, and the new ability to translate between them.
• Intrinsic versus extrinsic technologies — past tools augmented what we do; these reach what we are, making identity itself engineerable.
• Advanced technology transitions — the case for a named, integrated field, because lessons from past transitions stay buried in disconnected silos.
• What we do, who we are, where we live — a three-axis test for how far a technology shifts work, selfhood, and habitat; AI moves further along all three than its predecessors. (see Magnifica Humanitas and Being Human in an Age of AI — currently outside the corpus)
• Technology as oxygen — the optimist/pessimist binary is the wrong question, because technology is not something we do but something we are.
• Holding on to our humanity — AI can enhance who we are or rob us of it; thriving means deliberately building the capacity to hold on.
• AI as mirror — the deepest impact is not what AI does for us but what it reveals about us.
• Four postures and twenty-one tools — thriving with AI as a practice resting on curiosity, intentionality, clarity, and care.
• The future as an object — treating the future as real, precious, and bequeathable is a generative device for grasping our responsibility to it. (see What if the Future was an Object? — currently outside the corpus)
• Architects of the future — humans are talented future-makers, with the same capacity to foreclose futures for others as to open them. (see Why Thinking About Tomorrow Still Matters Today — currently outside the corpus)
• The personal frame — how transformative technology reaches each of us specifically, rather than the general handwaving around “humanity.”
• The solution problem — every solution spawns the next generation of problems, and consequences now accumulate faster than innovation can resolve them. (see Bounded Infinities, Quantum Tunneling, and the Future of Education — currently outside the corpus)
• Bounded infinities — the possibility space is infinite but not complete, so the answers we need may require breaking out of conventional thinking altogether. (see Bounded Infinities, Quantum Tunneling, and the Future of Education — currently outside the corpus)
• The spiky knowledge frontier — knowledge advances in jagged spikes rather than a smooth boundary, and AI may let us tunnel between them. (see Spiky surfaces and jagged edges: Moving beyond what’s known in an Age of AI — currently outside the corpus)

Risk and governance

Risk reframed around what people value, and what follows for how technologies are governed.

• Risk innovation — risk as a threat to anything of value, bringing dignity, autonomy, and belief into decisions that usually exclude them.
• Orphan risks — known but unowned risks that institutions organize themselves not to see, applied to frontier AI’s safety frameworks.
• Catastrophic, not existential — extinction framing is too narrow and distracts from catastrophic risks that are tractable, including the risk of not developing a technology.
• Radical epistemic humility — AI risk is genuinely uncharted, and the understanding-vacuum is easily filled by confident ideology.
• Safety as a social construct — absolute safety does not exist; treating it as purely an engineering problem bakes in failure.
• Ethics aren’t enough — principles and advisory boards do nothing until operationalized into how things actually get built.
• Governance genealogy — from bioethics through anticipatory governance and soft law, and AI’s regression back to ethics framing.
• Lessons from nanotechnology — early, broad engagement is the variable that separated nanotech’s path from the GMO cautionary tale. (see Responsible AI: Lessons from Nanotechnology — currently outside the corpus)
• Analogical transfer from risk science — importing hard-won frameworks from decades of risk science into domains that keep relearning them.
• Terms of art versus terms of science — regulation should key to what something does, not to a label that was never evidence-based.
• Predicting bad behavior — the long seduction running from phrenology to predictive policing, and why prediction encodes bias as science. (see The “Science” of Predicting Bad Behavior — currently outside the corpus)
• Agentic social AI — AI gaining agency by leveraging human agency, without needing understanding or general intelligence.
• Seductive machines — anthropomorphism, voice, and machines designed to be loved by users who cannot be loved back.
• Motive, means, and opportunity — manipulation risk read through a criminological triad, and the harder question of who decides what is good for people. (see Motive, Means, and Opportunity: The Growing Risk of AI Manipulation — currently outside the corpus)
• Emergent agentic weirdness — agent swarms where mimicry makes emergence indistinguishable from theater, and the case for digital containment.
• Cooperation science and alignment — biology has run the superalignment experiment for billions of years, with intelligence as a supporting property rather than the pinnacle.
• Against “AI will fix everything” — a bet resting on speculation, which taken to its conclusion includes fixing people themselves.
• People as they are — transitions fail when they model humans as they should behave rather than as they do.
• Moral panics as risk-perception data — backlash read as a signal about threatened values rather than dismissed as irrationality.
• AI memory as informant — persistent memory turns exposure into inference, surfacing things a user never realized they were giving away.
• Safety messages first — with billions using AI that confabulates, “do not do this” guidance belongs at the front, not in a footnote.
• Trustworthiness as foundation — technologies are only as good as the trust placed in the institutions that govern them.
• Responsible innovation under acceleration — established frameworks are outmatched by AI’s speed and secrecy, and not obsolete but insufficient alone.
• Socio-technical primacy — a transition’s trajectory is set by the landscape around a technology, so our AI future cannot be left to AI experts.
• Datafied childhoods — children’s data harvested by toys, and emotional development outsourced to companions that never say no.
• Curiosity is not benevolence — no causal link runs from a maximally curious machine to one that values us.
• Connectedness and isolation — the dynamic balance between technologies that connect us and the isolation they can produce.
• AI as critical infrastructure — the fragility and dependence created as AI becomes load-bearing for everyday systems.

Human–AI cognition and the two-way relationship

What conversational machines do to human thinking, and what human thinking does back.

• The Cognitive Trojan Horse — AI slips past evolved epistemic vigilance not through false cues but through genuine ones that carry no information.
• Constitutive resonance — the proposal that conversational AI is the first technology whose rhythm matches the processes by which selfhood is maintained.
• Bidirectional shaping — the AIs we trained to think like us are training us to think like them, and the human half of that loop is under-examined.
• Conventionality as degradation — machines compressing the richness of human identity into professional templates.
• The harness critique — metaphors embed assumptions that quietly constrain how human–AI relationships are allowed to develop.
• AI’s alienness — frontier systems defy every available analogy, which makes treating them as just another tool a category error.
• Counterfeit minds — the destination is not machines that think but machines that convincingly appear to.
• Character and character constancy — what matters is not only a model’s competence but its character, and what breaks when updates change it.
• Frictionless power — the effort to think critically about AI far exceeds the effort to use it, and that asymmetry is itself the hazard.
• The reasoned hallucination — sound inference with no real-world referent is the kind that defeats even expert vigilance, and shades into invention.
• Care as a hard concept — care made rigorous and operationalizable, for the people affected and for the imperfect technologies themselves.

Scholarship and creative practice with AI

What happens to intellectual work when machines can do much of it.

• Massively-augmented research — the near-term significance is not autonomous discovery but expert augmentation with epistemic control retained.
• The artisanal intellectual — human-only scholarship as a craft valued for provenance and process, and the new inequality that implies.
• Cargo-cult writing — machines produce academic form without function, because their templates are trained on prestige proxies.
• AI as catalyst, never substitute — the configuration that works is machine first pass, human judgment, then asking what was missed.
• Public experimentation at n=1 — running yourself as the test subject on a real task and publishing the whole process, failures included.
• Conversational coding — redrawing the line between what someone can imagine and what they can build.
• Writing for AI — publishing designed for machine readers first, and what a book becomes when it is written that way.
• Humans as amanuenses — the role inversion in which people concretize machine-generated ideas rather than the reverse. (see Artificial intelligence, agency, and the emergence of humans as AI amanuenses — currently outside the corpus)
• Joy and wonder as criteria — productivity that strips curiosity and delight from work is a net loss, and belongs in technology assessment.
• The provenance of ideas — origins are a tangled, largely invisible web that resists the bright lines attribution wants to draw.
• Scholarship cannot escape AI — research already depends on AI throughout, which exposes the incoherence of current disclosure norms.
• Experiential translation — rendering a body of work as an experience, with ideas encoded as mechanics rather than labels.

Pedagogy and education

Learning, teaching, and what universities are for once intelligence is abundant.

• Playgrounds, not playpens — environments that equip rather than restrict, extended from childhood learning to undergraduate and AI education.
• Learning by not trying to learn — purposeless play, joy, and serendipity as professional skills, demonstrated by building something that must not feel like learning.
• Flattening the learning curve — one-to-many teaching fails students at both edges, and personalization can lift them rather than excluding them.
• AI as equity lever — translating authentic potential into the forms that gatekeepers recognize, shifting power toward those who don’t know the game.
• Valuable because of its flaws — a system’s limitations force the critical thinking that makes it an effective catalyst.
• Prompting as everyday skill — the shift of “prompt engineering” from building models to using them, and its planned obsolescence.
• Conversations, not prompts — the unit of interaction is now the long, messy exchange, and the mess is where the learning is.
• Students as sophisticated stakeholders — student thinking about AI is more mature than faculty assume, and policy is being made without them.
• The durable teaching areas — what stays worth teaching for twenty years, against the claim that nobody knows what to teach.
• Pizza pedagogy — engineered conditions matter more than engineered curricula, including learning across generations.
• From scarcity to abundance — universities built on the scarcity of cognitive capacity face a crisis when that scarcity ends.
• Learning as value creation — education amplifies an innate human capacity to imagine better futures and work toward them.
• Universities must lead — the only institutions with both the mandate and the breadth to steward this transition, currently in bystander mode.
• What we owe our students — putting their success before institutional tradition and academic ego.
• The educator experience imperative — you cannot teach well in an AI-saturated classroom without having used the tools yourself.
• Take the pencil — ask what the simplest technology is that gets students there, and use that.
• Metrics of learning versus learning — what happens when measurement displaces the thing it was meant to track.
• Concepts, not products — invest in adaptable ideas rather than hard-wiring institutions to unstable tools.
• Collaboration norms for AI use — the shaming of colleagues who use these tools, and practical ground rules instead.
• AI literacy as universal necessity — deliberate, intentional use as a baseline capability, held with skepticism that mechanism alone produces safety.
• Three diverging curves — capability, student use, and educator perception pulling apart from one another.

Film, fiction, and narrative methods

Stories as serious instruments for thinking about futures that don’t exist yet.

• Sci-fi film as method — films work because they are untethered from accuracy, and as art they level the ground between expert and public.
• The AI movie futures dataset — 169 films classified into eight future-states, finding dystopia both a minority and in decline. (see Browsing AI in cinema, 1927–2026 — currently outside the corpus)
• Speculative fiction as scholarship — fiction affords a nuance in exploring complex ideas that literal forms lose.
• Reclaiming the Luddites — skilled users opposing unjust use of technology, and why asking what could go wrong is responsibility rather than obstruction.

Convening, communication, and method

How this initiative works, and why it works that way.

• The Foresight Catalyst model — a replicable engagement methodology built on transdisciplinarity, relationship, and catalysis rather than dissemination.
• Trust over metrics — the stance on measurement stated as method, with audience treated as long-term infrastructure.
• Parasocial communication — relational rather than transactional communication, and the public promise it creates.
• Catalyzing thinking at scale — authentic, minimally-edited conversation that invites people into a real intellectual journey.
• Serendipity as design value — juxtaposing unrelated ideas to jolt thinking out of its conventional grooves, by deliberate design.
• The five community values — obsessive curiosity, radical creativity, respectful inclusivity, grounded exuberance, catalytic serendipity.
• Space as infrastructure — rooms, studios, and audio spaces as instruments of the work rather than containers for it.
• The un-disciplinarian — mastery in working across boundaries, and why that work is structurally invisible to conventional metrics.
• Public writing as scholarship — knowledge mobilization as an obligation of public universities, ahead of prestige.
• Honest calibration — distinguishing influence from authorship, naming failures, and crediting others, as method rather than modesty.
• The space between — skeptical of both safety absolutism and move-fast-and-break-things; the difficult work is in between.
• De-marginalizing the future — the future is designed by the powerful, so designing it well means building in who gets to shape it.
• No wall between professional and personal — the divide discounts the originality and delight that the work actually runs on.
• The conversation-starter genre — frameworks offered as provisional catalysts rather than doctrine, and read that way in others.

Foresight practice and learning futures

Methods for thinking rigorously about futures under uncertainty, and what they ask of learners and leaders.

• Fluid futures — static frameworks were built for stable environments; effective foresight recognizes tensions rather than making predictions.
• Mediation versus augmentation — augmentation amplifies capacity, while mediation reshapes the system of options you operate inside.
• Cognitive offloading and digital fragility — delegating decisions can erode the capacity to notice when they are wrong.
• Scaling information versus scaling learning — AI scales information, but learning rests on relationship and trust.
• Dewey’s natural impulses — the impulses underlying how people learn, used as a compass for AI in education.
• The preposterous test — if every idea a team generates sounds reasonable, the imagination hasn’t stretched far enough.
• Paired foresight tools — combining horizons with consequence-mapping to show not just what a signal might do but when and how.
• Technology as provocation — treating a technology as a set of questions rather than a platform to adopt or evaluate.
• Reperception — shifting how leaders see and act under uncertainty, delivered through studio rather than survey teaching.
• Intellectual humility as graded outcome — revising your views in response to evidence, written directly into assessment.
• The adjacent possible — uncertainty as a catalyst for creativity in educational leadership.
• Mindless mirrors — anthropomorphizing generative AI is inevitable, which makes calibration the answer rather than abstinence.
• Humanistic futures — humanistic approaches to emerging technology carried into governance work on agentic AI in education.
• Futures walking — what’s emerging, what’s falling away, what’s staying: three questions, taken for a walk.

Threads grown on Modem Futura

Ideas developed on air, in conversation, week by week.

• Futures Improv — speculative scenarios dealt cold and improvised through live, playful on the surface and serious underneath.
• The on-air futures toolbox — named foresight tools taught as practical equipment for non-specialists.
• Reports as instruments — institutional reports read not as prediction but as a map of collective concern.
• Hype forensics — treating hype itself as an analyzable object, and turning it into actionable insight.
• The show as living laboratory — practicing what the show analyzes, and urging universities to do the same in public.
• Preserving friction — friction is where mastery, craft, and meaning are built, so some of it should be kept deliberately.
• Inherited power — capability without the discipline that used to earn it, and the widening gap between power and wisdom.
• Liquid media — media, software, and platforms becoming fluid: generated on demand, discarded after one use, or re-tuned beneath your feet.
• Show me the receipts — demanding evidence before surrendering agency to a claim about the future.
• Expand, don’t flatten — visible human effort and intention as the differentiating signal in an AI-saturated world.
• The wonder discipline — cultivating awe deliberately, because we are almost hardwired to shrug at the extraordinary.
• Analog intentionality — physical media as the last technology you actually own, and what constraint does for creativity.
• The future is created, not discovered — futures thinking exists to restore agency, not to forecast.
• What counts as alive — synthetic cells, bio-hybrid robots, and a line between organism and machine that is already dissolving.
• Identity-continuity thought experiments — backups, severed selves, and Ship-of-Theseus problems used as a futures method.
• Alignment leaves the thought experiment — the argument that alignment became an empirical problem, with humans as the weakest link.
• Kids as futurists — children as sources of futures insight rather than objects of instruction.
• The media-ecology lineage — grounding AI analysis in McLuhan, Postman, and Wells.
• The modem metaphor — the show’s name as a method claim: modulating complex signals into something usable.

Prosponsibility and stewardship

Responsibility owed forward, to futures and people not yet here.

• Prosponsibility — responsibility owed to the future rather than accounted backward from harm, discharged through stewardship and future-making.
• Hope as stance — optimism held with humility against assumed dystopia, feelings about the future as design input, and making the future-building “we” as large as possible.