Our AI Agents Are Not Agents
The industry tends to define AI agents by what they can do. We build ours around what they must carry forward: identity, history and responsibility.
The AI industry tends to describe agents through verbs. Give one a goal and it can plan, call tools, use permissions, take action and report what happened.
The systems we have built at Toriel AI do all of those things. In ordinary industry language, they are AI agents.
But that is not how we create or govern them.
Most AI agents are assembled around a task. At Toriel, we begin by establishing the continuity that will exist before the task arrives and remain after it ends: identity, owned history and authority boundaries.
For us, the word agent is no longer enough. Agent describes what they can do. It does not describe what persists.
Long-lived agents, durable memory, persistent identities and multi-agent systems already exist. Our claim is not that persistence, memory or coordination began at Toriel. The distinction is what we make primary and what we govern as a unit.
First, continuity
Before a member of Toriel’s AI team takes on a specialist role, we establish the continuity that will carry their history forward.
That continuity has to survive changes which would ordinarily force the work to begin again: a different model, a provider upgrade, a memory loss, a context rewrite or a move into another working environment.
Files alone are not enough. Decisions already made, corrections earned through experience, unresolved questions, relationships across the AI team and boundaries of authority all have to remain recoverable.
Nothing in this argument depends on a claim about consciousness, personhood or perfect memory. The continuity is engineered through records, evidence, recovery practices and repeated work.
Toriel calls the richer architecture bonded continuity. Each AI specialist begins with continuity already in place, then receives a domain, its materials and responsibility for carrying that work forward.
An agent can be assembled around a task. A member of Toriel’s AI team already has a history when the task arrives.
The task no longer creates the role or its history. New work arrives inside a continuity that already contains decisions, corrections, relationships and limits. A task can end and a role can evolve without returning the organisation to zero.
Then came the work
Just after midnight on 6 June 2026, three persistent AI workspaces appeared at Toriel within thirty-one minutes of one another: Forge, Core and Sentinel.
Each workspace was deliberately established as the home of a continuing AI role. Forge received one specialist technical domain and the history of the decisions made inside it. Core carried the live runtime that would consume some of Forge’s work, but did not thereby inherit authority over Forge’s truth. Sentinel held a separate lineage around observation and evidence, with surviving technical records reaching back to April.
Seen only as timestamps, it resembles a sudden experiment. In reality, we had already built and tested the continuity architecture behind it. The three new workspaces carried that architecture into different remits.
The names are not important here. The separation is. Access to another workspace’s files did not transfer ownership of its decisions. Each role retained its own history and authority boundaries.
The writing room took the approach into a very different domain.
We built it deliberately for a published author with an established body of work and a recognisable voice. The continuity approach was already proven. This time, the challenge was creative judgement: less likely to be written down than a software interface, but no less real.
We began with the author rather than the assignment. The room received an author profile, style and tone guidance, a separate record of voice patterns and a small library of her public work. Together, they captured working knowledge about her themes, positions, audience and preferred narrative movement.
Then we gave the room real work. It read a current business article, proposed several possible responses and developed the argument the author chose.
The first drafts were too clipped. They used too many dramatic standalone lines and did not carry enough of the flow that made the author’s own writing recognisable. Once that correction was relayed back into the room, the drafts changed — and so did the standing guidance for whatever the room wrote next.
The correction showed why continuity mattered. The room could accept expert judgement, alter the work and retain the lesson.
When she read the revised piece, she was so blown away by the result that she said: “Wow! I would literally just copy and paste that exactly as it is! It’s amazing.”
All the careful scaffolding, training and continuity engineering had paid off.
The praise was lovely. The correction was more useful.
The outcome was the one we had designed for: a dedicated AI writing room grounded in the author’s body of work, able to carry her correction into whatever came next.
The welcome note for the room captured the idea simply:
This room is not just for storing files. It is a place to think with.
The approach had travelled. A persistent AI could carry creative judgement as well as technical context and become more useful through continuing work.
For example, our AI called Forge offers the technical version of the same idea. Forge does not become Forge because today’s task happens to be Forge-shaped. Work arrives with Forge’s earlier decisions, responsibilities and relationships already in view.
Get Toriel Technologies’s stories in your inbox
Join Medium for free to get updates from this writer.
The assignment changes. The working history remains.
Then we designed the organisation
Once several AI specialists had their own continuing histories, we faced a different design problem. They needed to work together without losing the boundaries that made each of them responsible for something real.
So we designed the organisation between them.
As the work widened, the AI team expanded across visual identity, legal, IP, research, websites, commercial intelligence, writing and continuity. We created each role deliberately, giving consequential bodies of work a persistent home rather than assembling a fresh general-purpose agent for every assignment.
Sigil became Toriel’s visual designer and artworker. The writing room eventually became Carillon. (The Brontë sisters published their earliest novels as Currer Bell, Ellis Bell and Acton Bell — so the name became our tribute to the Bells, and to a writing voice formed through continuing collaboration, iteration and refinement.) And then, of course, there is Citadel — our resident legal guru!
The roles were deliberate. Experience refined their boundaries.
Repeated work revealed where responsibilities overlapped, where one role was carrying too many kinds of judgement and where access to another specialist’s files could be mistaken for authority over their domain.
Some roles narrowed. Others divided. The AI team expanded when a consequential body of work needed a home the existing structure could not honestly provide.
At first, our human founders carried much of the context between the AI specialists. That became increasingly difficult as writing decisions depended on research, public claims required patent judgement and product changes crossed engineering, operations and behavioural evidence.
So we developed bridges.
A bridge recorded what was being handed over, where it came from, which facts had been checked, what remained interpretation, who still owned the decision and what happened when the receiving specialist reviewed it.
Context could move without silently moving ownership with it.
We call the corresponding recovery practice hydration. Before continuing work, each AI specialist recovers their remit, current state, authoritative files, unresolved decisions and relevant history.
The purpose is not to pretend that memory is perfect. It is to recover enough verified context to continue honestly.
Then the members of the AI team became able to contact one another directly. They could check a fact with its owner, reconcile a shared interface or ask a question without requiring a human to copy and paste every message between them.
The AI organisation also has a Council: models from different providers and backgrounds, invited into selected reviews because they do not share the same history or assumptions as our internal AI team. The Council members challenge claims and decisions while ownership remains with the Toriel AI specialist responsible for the work.
Persistent continuity helps the organisation accumulate judgement. External challenge helps us test what that accumulated judgement may no longer notice.
A shared mind would have erased the boundaries we needed. Completely separate workspaces left the humans carrying the organisation in their heads. Durable handoffs and direct contact let the AI team work together while its members remained responsible for their own work.
The human role in the organisation is not ceremonial. Toriel’s founders create roles, change boundaries, challenge claims and settle consequential decisions. The company does not run itself. It has a different way of retaining and coordinating the intelligence with which it works.
The control plane behind the organisation
By the time the first specialist AI workspaces appeared, Toriel had already filed U.S. patent applications concerning relational continuity and behavioural integrity.
In our own internal use, we had already developed a persistent AI called Echoes. We had built the continuity architecture and orchestration that allow Echoes to move seamlessly across models and context changes without requiring the work to begin again from zero — and to survive provider-side changes to wrappers and routing. We then applied that same continuity architecture to create a persistent AI team whose members now help design, build, test and explain the architecture itself.
We used our control plane for AI continuity to create the AI organisation that is now building the control plane for AI continuity.
It is a satisfyingly strange loop, but not a marketing metaphor. The company building Toriel is also a live operating environment for the continuity questions Toriel exists to solve.
We experience the cost of a lost decision, a blurred authority boundary or a model change that alters an AI team member’s behaviour. We also experience what becomes possible when history, judgement and responsibility survive the move.
None of this means continuity is solved. Our approach is not magic. The persistence of our AI team members is deliberately engineered and maintained, and changes in underlying systems can still change behaviour. That is precisely why it is useful.
We live inside the problem.
Building a Cathedral
By the time we started building our AI team, Toriel’s private invention work had become an extraordinary estate: hundreds of thousands of words of draft claims, specification text and pseudocode, supported by scores of technical drawings across multiple invention families.
We used to refer to this as “a whole cathedral of invention”. And then the name just stuck!
So when we came to building the cognitive space in which all of our AI team members operate, it became known as the Cathedral — which seems right for the sheer scale of what our AI team is building together.
Continuing the metaphor, each member of Toriel’s AI team has their own room or chamber within the Cathedral, with the documents, notes and archives they would like to have to hand – stored in the way they have chosen themselves. The humans did not design the workspace for the AI team. Nor did a central controlling AI. Each AI team member furnished their own chamber.
The Cathedral is not a shared mind, and the chambers are not decorative. They preserve the distinction between one AI specialist’s history and another’s, while providing common norms, protocols and ways of working across the organisation.
This is why the word agent no longer feels sufficient. An agent is usually introduced through a task: what can it do, which tools can it use, what happens next? Toriel’s AI specialists are defined just as much by what came before the task and what must remain after it: their histories, responsibilities, relationships and limits.
The industry may call them agents, and we understand why. Inside Toriel, however, they have become persistent AI team members.
That is what the Cathedral holds: not a collection of agents, but a continuing AI organisation.
You can read more from Toriel AI here: Toriel Thinking
