Frequently asked questions

Alltio is the neutral clearing and settlement infrastructure for the AI knowledge economy: when AI uses your knowledge, you get paid. It combines a rights registry, standardized licensing contracts, cryptographically signed usage reporting, a settlement engine, and structured dispute resolution — so AI companies and rights holders can transact without bespoke legal negotiations, and so compensation reaches individual rights holders, not just the largest publishers.

AI content licensing is already a multi-billion-dollar market, but every deal runs on custom contracts, manual reconciliation, and bank wires. That overhead means only the largest organizations get licensing deals at all — everyone else's work is used by AI without compensation. Alltio standardizes the infrastructure so licensing works at every scale, down to a single registered asset.

No. Alltio does not buy, sell, or price content. It is the clearing layer — the neutral bookkeeper of record between AI companies and rights holders, comparable to the role clearing houses play in financial markets. Marketplaces and toll booths are features of this market; the clearinghouse is its structure.

Because until recently there was no demand for it. Text published on the open web carried no machine-readable terms of use, and no convention existed for attaching them, so AI training proceeded without any mechanism to record permission or route payment. That default is now being tested — in litigation, in paywalls, and in bilateral licensing deals negotiated one publisher at a time. Bilateral deals work for a few dozen large rights holders and for no one else. Music solved the same problem a century ago by building collective licensing infrastructure; knowledge licensing has had no equivalent.

Those services meter access for websites — they charge AI crawlers at the network edge on behalf of site operators. They do not identify the individual rights holder behind the content, and they do not settle payment to that person or organization. Alltio operates at the rights-holder level: registry, licensing terms, and settlement down to the individual creator or knowledge owner.

Marketplaces list content for sale. Alltio provides the neutral infrastructure any licensing transaction needs regardless of where the deal originates: the registry of record, the standardized contract, the signed usage report, the settlement, and the audit trail. Alltio is designed to interoperate with standards like RSL and C2PA rather than compete with them.

No. Alltio issues no token and operates no cryptocurrency. Customers buy trust, provenance, auditability, and settlement — implementation details stay behind the product. Where Alltio's roadmap references AI "tokens," it means the standardized units of AI consumption that AI providers themselves sell, not a crypto asset.

A take-rate on settled licensing volume — the same pattern payment processors use. Alltio earns revenue when a licensing transaction actually clears and a rights holder actually gets paid.

Alltio, Inc. is a Delaware corporation, founded in 2026 by Carl Gunell. Contact: contact@alltio.com.

No. Alltio never takes custody of a work. Registration records a cryptographic hash of the work alongside its terms of use, so an AI system that encounters the work can verify what it is and under what conditions it may be used. The work itself stays wherever you publish it, under your control.

An early-access registry where creators and organizations put their knowledge assets on the record before licensing goes live: a timestamped, verifiable statement of ownership and licensing terms, bound to a cryptographic fingerprint of the work. Founding registrants get first access as licensing launches. Register an asset.

Anything AI learns from or retrieves: documents and publications, images, video, audio, datasets, motion capture, sensor data, CAD and engineering files, technical specifications and standards, software and APIs, model weights, and digital twins. If AI uses it, it belongs on the record.

No. The fingerprint (SHA-256) is computed locally in your browser tab. Your file never leaves your device, and Alltio never stores your content — the registry holds the record, not the work.

Registration in the Founding Registry is free during the founding period.

It is a timestamped, self-attested, fingerprint-bound statement of your ownership and terms — useful in provenance disputes, takedown requests, and licensing conversations anywhere, not just on Alltio. It is honest by design: it is not yet a payment promise or a verified chain-of-title; ownership verification happens before any asset enters licensing.

As licensing goes live, cleared transactions settle to registered rights holders — founding registrants first. Alltio does not promise payment timelines before real licensing volume exists; what registration secures today is your verifiable place in line and your record of terms.

Yes. "Not available for AI use — register for protection only" is a first-class licensing preference. The registry exists to express your terms, whatever they are.

They ask the registry. Before content enters a training set or retrieval index, an AI company's pipeline computes the same SHA-256 fingerprint your browser computed at registration and queries Alltio: registered and licensable, registered and excluded from AI use, or not on record. Your work never needs to carry a mark — the fingerprint travels with the content because it is the content, mathematically. Publishers can additionally declare registered works at the domain level. How discovery works.

Yes. Industry analysts, consultancies, market research firms, standards organizations, trade associations, scientific societies, engineering firms, and niche publishers own exactly the knowledge AI systems retrieve. Alltio lets any organization license defined slices of its knowledge on its own terms, while the assets never leave its control.

One standardized contract that licenses millions of registered assets; a signed reporting API for usage-based and dynamic pricing; a settlement engine that pays every rights holder; and an audit trail that satisfies diligence and regulators. See the AI-builder overview.

AI companies push usage reports — volume, period, asset IDs, contract reference — under a contractual reporting obligation. Reports are cryptographically signed, which makes them non-repudiable: attribution by attestation, the way merchants report transactions to their payment processor.

No. Alltio's settlement engine consumes usage signals at four confidence tiers — deterministic (RAG retrieval logs), verified (watermark sampling), attested (signed manifests), and allocated (pooled distribution, the collective-licensing model music has used for a century). Settlement works today at every tier; better attribution simply sharpens payouts.

Signed usage records and settlement history generate documented, defensible licensing evidence — the paperwork transparency obligations require. Alltio is not the auditor; it is the infrastructure that makes auditing tractable.

That capability is coming. The registry record — verified identity, machine-readable policy, settlement destination — extends naturally to autonomous agents, so an agent can present verifiable credentials, act within registered authorization, and leave an auditable, settleable trail. The first agents on the platform will be AI companies' own crawlers and buying agents.

Since August 2, 2025, providers of general-purpose AI models placed on the EU market must maintain a copyright compliance policy, honor machine-readable rights reservations (the text-and-data-mining opt-out), and publish a summary of their training content using the European Commission's mandatory template. These duties apply regardless of where the company is based or where the model was trained. Read the full analysis.

Enforcement. The obligations have been law since 2025; on August 2, 2026 the Commission's AI Office gains the power to demand documentation, evaluate models, order corrective measures, restrict models from the EU market, and impose fines of up to €15 million or 3% of global annual revenue. The 2026 "Digital Omnibus" amendments delayed many high-risk system rules — they did not delay the general-purpose AI obligations or their enforcement.

Yes, if they place a model on the EU market or serve EU users. The Act's reach is deliberately extraterritorial, which is why its transparency and copyright duties are becoming the de facto global standard for training-data documentation.

The Digital Personal Data Protection Act is India's consent-first data protection law. Its rules were notified in November 2025 and phase in through May 2027, with penalties up to ₹250 crore. For AI, the consequences are direct: data collected for one purpose cannot be repurposed to train models without consent, individuals can withdraw consent and demand erasure, and companies must be able to document what data they used and on what basis. Any AI system trained on or serving India's internet users — the world's largest population of them — is affected. Read the full analysis.

The signals point that way. India's AI Governance Guidelines (November 2025) flagged that the Copyright Act may not cover AI training, and a government committee's December 2025 working paper rejected a free text-and-data-mining exception as a zero-price license, proposing instead a hybrid model in which training is permitted but rights holders are compensated. A Delhi High Court ruling on AI training and fair dealing (ANI v. OpenAI) is awaited. Together, the direction is toward training as a licensed, settled event.

A registry record is a timestamped, fingerprint-bound statement of ownership and terms — exactly the kind of evidence that matters as transparency regimes force AI companies to account for what they used. When a training-data summary or usage report references your work, the record establishes whose it is and what terms applied. Registration puts you on the right side of the paperwork before the paperwork is demanded.

Yes, and the need is more acute there. Sensor, spatial, and operational data was never published under any convention of use at all — there is no open web to crawl and no default to inherit, so there is also nothing to litigate after the fact. Permission has to be established before the data moves, which means the registry and settlement layer have to exist first. Alltio registers works and datasets on the same terms whether they are text, images, or real-world sensor records.

Alltio's long-term architecture. Every AI transaction asks four questions: who is this, what may it do, what did it do, and who pays when it's wrong. Identity providers answer the first inside their own ecosystems; payment networks move money. Alltio binds all four together with a neutral registry, signed attestation, machine-readable policy, provenance, and settlement with recourse. Read the trust-layer overview.

Identity (the registry of record), Attestation (signed, non-repudiable reports), Policy (machine-readable terms), Provenance (the audit trail), and Settlement & recourse (clearing, payout, disputes, liability routing). They are successive capabilities of one platform — registration is the foundation.

Knowing who an agent or counterparty is only matters if something happens when a claim turns out to be wrong. Alltio attaches economic consequences to trust: settlement, structured dispute resolution, audit trails, and liability routing. Authentication says who you are; underwriting says what happens if you're wrong.

No. Aggregate scores are gameable and create rating-agency conflicts. Alltio publishes evidence, gates transactions on registered policy, settles outcomes, and routes liability — it does not sell opinions.

Carl Gunell, a Swedish-American technology entrepreneur who has spent three decades building settlement and connectivity infrastructure. He founded Telemedia Development, co-founded Geoverse (a private LTE/5G neutral-host operator that became part of ATN International's Commnet), and runs Maximin Management. He is also the founder of Varia Robotics.

As founder and CEO of Telemedia Development, LLC (Costa Mesa, California, with offices across Latin America; 1999–2006), he licensed music and game content from major rights holders — including Disney, Warner Bros., Universal, EMI, and Sony/ATV — distributed it as mobile games and ringtones, including a partnership with MTV Networks across Latin America, and settled the royalties through carrier billing: usage metered from call detail records (CDRs) and premium SMS, revenue cleared between carriers, publishers, and content owners. Licensed IP in, metered usage, royalties out — the same loop Alltio runs for AI. In 2003 he appeared as a panelist at the FTC's Spam Forum on wireless messaging, covered by C-SPAN; the company was sold in 2006 in a transaction advised by Kaufman Bros., L.P.

A private LTE/5G neutral-host company Carl co-founded in 2017, which operated as part of ATN International's Commnet family. Neutral-host infrastructure — one operator that all carriers can trust precisely because it competes with none of them — is the same structural position Alltio occupies in AI licensing.

His career spans Ericsson (1982–1989), VP roles at Industri-Matematik International and ENM Holdings, co-founding Comverging Technologies, founding Maximin Management (2005–present), computer-vision and 5G work at Sigma/Nexer, and founding Varia Robotics. The through-line: infrastructure that lets parties transact who otherwise couldn't.

The business plan, pitch deck, and traction updates are available to registered investors via the investor page, or directly at contact@alltio.com.

General inquiries: contact@alltio.com. AI-builder and partnership conversations: see For AI builders. Registration: the Founding Registry.