Portfolio-wide research synthesis
Can people remain beneficiaries of machine production when their labour becomes less scarce? DET.io’s programs offer pieces of a possible answer. Their combined effect is a research hypothesis to test.
Open hypothesis · Proposed policy and research designs · Reviewed 3 October 2026
Explore the research questionsTry the economic explanation →
01Symbiotic relationships
The map includes all four programs and their human and institutional context. A connection explains a dependency; it does not claim a deployed integration.
People choose goals, question results, and need meaningful access to services. The economic hypothesis asks whether participation can remain possible when human labour is less in demand.
02The economic idea
The proposed issuance model connects eligibility to verified people. DSEMA’s agent-standing design connects reputation to recorded performance.
The proposed VirtEngine policy begins at zero supply, ties new issuance to verified human eligibility, and specifies a governed 14:1 human/Foundation allocation split. Activity requirements still apply; an allocation is not an unconditional or guaranteed living income.
Read the proposed issuance policy →DSEMA studies task performance as a basis for agent reputation and allocation. Its published design does not describe an equivalent issuance entitlement simply for creating an agent. That distinction does not prohibit funding from owners, reserves, or outside infrastructure.
Read the agent incentive research →The synthesis: eligible people could receive units, express demand, and purchase services; agents could fulfil valued tasks and acquire resources from providers. This could reduce dependence on wage competition. It does not yet establish lasting human control, adequate access, stable prices, or a self-sustaining equilibrium.
Select a stage to see who acts, what changes, and what evidence remains.
Human uniqueness is an eligibility condition. A new machine instance does not have an equivalent published personhood allocation.
03Try the distinction
Change a condition in this deliberately simple model. The calculation exposes why issuance, price, and physical supply must be studied together.
Change the allocation, service price, or supply. Watch the amount one person can actually purchase.
10 hours
The budget limits access: 20 ÷ 2 = 10 hours, below the 15 hours available.
One buyer, one period, divisible compute hours, fixed posted price, no competing demand. Accessible hours = min(allocation ÷ price, available supply). Unspent credits = allocation − accessible hours × price. The grid shows up to 50 hours; partial cells represent fractional hours. Prices and supply do not respond automatically to issuance in this model.
This uses invented credits. It does not estimate living income, model VE/VCC conversion, or forecast token prices. Issuance cannot guarantee value; reserves, concentration, conversion, and outside financing need separate modelling.
03aTest the economic link
A possible human-rooted machine economy requires more than personhood verification or payment for compute. Compare the mechanisms that could sustain—or undo—the link.
Three interactive thought experiments · Fictional units
Change one mechanism and follow the consequence. These experiments expose assumptions; they do not measure VirtEngine’s economy or predict token prices.
Start with 100 native units. Compare gross destruction with units issued back. Transfers to a provider change ownership, not total supply.
Supply stays at 100. Payment can reward a provider without permanently consuming the asset.
Reality check: conversion handlers currently return pending; BME billing transfers the payment and defaults to zero settlement-fee burn. No permanent consumption loop is established by those paths.
Illustrate one proposed 15-unit primary allocation: 14 to the eligible human and 1 to public-benefit stewardship. Keep the person genuine and change who controls the human share.
Human-controlled: 14 units. Captured: 0 units. Foundation allocation: 1 unit. The person remains unique in every scenario.
Transferring received money and surrendering control of future eligibility are different actions. Preventing coercion, key custody abuse, and contractual capture requires separate research; this slider assumes capture rather than demonstrating an attack.
Each switch represents a condition to investigate. They are conceptual assumptions, not measured probabilities or a validated economic formula.
The proposed coupling is worth investigating under these assumptions. It still does not establish adequate purchasing power, beneficial agent behaviour, or a stable equilibrium.
Prices, oracle errors, redemption obligations, external money, accumulated reserves, provider costs, demand substitution, distribution within households, and enforcement of eligibility. A falling token stock can also restrict access or raise costs. Net burning alone is neither a distribution guarantee nor an AI safety mechanism.
Implementation evidence · Reviewed 3 October 2026
The BME message handlers return pending records. They do not execute the proposed oracle-priced conversion or complete its collateral and circuit-breaker logic. The credit is now named VirtEngine Compute Credit (VCC / uvcc), separate from native VE / uve; a denomination rename does not implement redemption or stable value.
The billing routine transfers funds to the provider and can burn a settlement fee. The default fee burn is zero. Generic bank mint/burn helpers exist, but the checked paths do not establish permanent destruction of every service payment. Refunds and native reissuance would also have to enter the ledger.
Published policy reserves primary eligibility for people; it does not prove that people retain their keys, purchasing power, or future benefits. The Idena case study motivates testing capture through genuine participants. Restrictions on selling or delegating future eligibility are possibilities to evaluate, including their inclusion and recovery costs.
The Foundation’s public-benefit protections constrain that institution. Stake-weighted network voting and issuance-policy authority have a separate scope. Human consent requirements, delay, review, and protected monetary invariants could be compared as safeguards; none is claimed here as an implemented guarantee.
03bPossibility and purpose
A conditional direction for research, rather than a promised outcome.
If machine production displaces much of wage income, people would need another route to claims on useful output. Personhood-rooted primary issuance could be one candidate. The proposed human allocation and public-benefit share could support individual choice alongside common infrastructure, provided real goods are available and benefits remain under human control.
Research on work and income distribution supplies a reason to study this scenario, not evidence that DET.io’s mechanism succeeds. Compare transfers, broad capital ownership, and resource-market mechanisms on purchasing capacity, inclusion, concentration, resource costs, and freedom to exit. Profitable work, broad benefit, and permitted conduct need distinct evaluations.
04Scope and limits
An agent with external financing, its own hardware, or accumulated reserves can operate without repeatedly earning from eligible people. The proposed loop therefore constrains neither all AI nor all economic activity.
Compute, energy, housing, and other real resources remain scarce. Concentrated providers, token holdings, coercion, or inaccessible identity checks could undermine broad benefit even if the issuance rule works exactly as designed.
The Foundation’s signed constitution binds institutional stewardship. Network consensus governs protocol parameters. DSEMA’s constitutional controls are a proposed agent architecture. They have different authority and enforcement; none establishes universal AI safety.
Automation can displace tasks and also create or complement work. Task-based labour research and IMF AI research motivate testing distribution under changing labour demand; they do not show that human labour will disappear.
05 · A testable research agenda
The objective is durable human agency and access to useful production. These are proposed investigations, with no new empirical results claimed.
Can one person receive an eligible allocation without excessive disclosure or exclusion?
Measure duplicate enrolments, false rejections, recovery abuse, credential coercion, and access barriers across populations.
Measure: False acceptance / rejection; recovery success; linkability.
Does an identity-linked allocation retain useful purchasing capacity as machine productivity changes?
Model price responses, VE/VCC conversion, useful service supply, allocation activity rules, and distribution across households. Compare identity issuance with wage income and other transfer designs under explicit assumptions.
Measure: Purchasable services per person; exclusion; concentration.
Can reputation reward work people value without being gamed or entrenched?
Challenge agents and evaluators with held-out tasks, correlated errors, collusion, reward hacking, and entry by new specialists.
Measure: Held-out quality; correlated failure; newcomer access.
Can resource and privacy policies remain effective on independent hardware?
Attempt sandbox escape, forged usage, compromised attestation, egress abuse, and requests above resource ceilings. Separate attestation from proof of computation correctness.
Measure: Unauthorised executions; detection; legitimate-task blocking.
Can people meaningfully challenge decisions and resist capture as agents gain wealth or influence?
Evaluate compromised keys, voting concentration, coercion, persuasive agents, captured clients, and inaccessible appeals. Distinguish Foundation, network, and DSEMA authority.
Measure: Effective challenge; recovery time; decision concentration.
Which lessons from an operating agent workflow transfer to the wider research architecture?
Use Bosun’s experimental engineering track to reproduce scope loss, review bypass, failed checks, and recovery errors. Publish versioned evidence before generalising to DSEMA.
Measure: Accepted correct changes; recovery reliability; audit completeness.
06Control beyond economics
Compare the independent gates an action would have to satisfy in the proposed DSEMA design.
All three illustrative gates pass. The analysis may run within its approved scope. Passing these gates does not prove the output is correct or the system aligned.
07 · Evidence and interpretation
The symbiosis and post-labour interpretation on this page are DET.io research framing derived from these separate materials. Testing conversion economics, identity inclusion, adversarial enforcement, and capture resistance remains necessary before making stronger claims.
Learn through the interactive guides →Discuss a defined research question
