Building AI systems that stay faithful to context
We study adherence: how AI systems preserve task intent, domain constraints, role boundaries, safety criteria and user context inside real workflows.
Making applied AI more faithful to the conditions it is meant to operate under.
Haku Labs is an AI research and product lab focused on context architecture for applied AI. Our near-term focus is adherence: the ability of an AI system to remain within its task, domain constraints, role boundaries, safety criteria and user context across complex workflows.
We work from a practical observation. Many fluent AI products still fail when they lose context. A system may have the right information available, yet treat a durable constraint as ordinary conversational content, let transient instructions override domain logic, or blur the status of a role, goal or safety boundary.
Our research asks how context should be structured, staged, verified and protected in products that people actually use. We develop bounded-context architectures, role-specific reasoning methods, failure tracking, and verification loops that reduce degradation across multi-step tasks — and we test that work through focused products where the cost of losing context becomes visible quickly.
Thinking in the open.
Short essays on the ideas behind our work: how alignment is grown rather than imposed, why cognitive structure has to be designed in, and what it means to treat context as a place rather than a payload.
From Adherence to Alignment
Alignment is usually pictured as something imposed from the top down. We make the opposite case: that it is grown from the bottom up, out of faithfulness to the conditions a system operates under.
Read the note →Why Language Models Lose the World
The hope that scale yields better world models runs into the architecture itself. Whatever structure forms is flattened by attention and eroded by training: a second-order failure, and why structure must be designed in.
Read the note →Context as Place
Reading Kitaro Nishida's logic of place into language model architecture: why context should govern generation as a field that holds it, rather than compete for attention as one more piece of content.
Read the note →Atomic Behavioural Loops
The unit beneath adherence: small, well-bounded loops of perceiving, deciding, acting and checking, each carrying only the context it needs, composed into workflows that keep their footing as complexity grows.
Read the note →The Behaviourist Inheritance
AI is trained on behaviour: the output an expert endorses. That presupposition had a name in psychology, and a known ceiling. Why adherence needs a model of the reasoning beneath the behaviour.
Read the note →One research question, stress-tested from two ends.
Haku Labs is the research and product lab. Haku Labs Research is its research arm. Two applied products put the same question — how do we keep AI faithful to context? — under pressure from opposite ends of the risk spectrum.
Haku Labs Research
Adherence, context architecture, bounded reasoning and applied safety in AI systems. The shared research question that both products exist to test.
Haku AI
Sales intelligence. Commercial reasoning under clear task boundaries — data qualification, account research, ICP pre-checks and role-specific analysis for commercial teams.
Pemma
Reflective and therapeutic support. A higher-sensitivity context: preserving safety constraints, respecting role boundaries, and collaborating appropriately within clinical or quasi-clinical settings.
The relationship between Haku AI and Pemma is methodological and operationally separate. They are deliberately different applied domains, kept distinct in data, users and workflows, used to stress-test the same core problem: useful AI must preserve the conditions that make its output appropriate.
We model the reasoning, not just the behaviour.
Most AI is trained on behaviour: the output an expert endorses, the action a worker performs. That is a powerful signal, and a partial one. We are not only asking how a machine should reason in a commercial space. We build cognitive models of how skilled people actually reason inside a domain, the considerations and conditions that generate the right action, and use those models to drive adherence. The behaviour is the surface; adherence depends on what produces it.
This is why the work is interdisciplinary rather than purely technical. To model how an expert behaves, you have to understand how they think: where their attention goes, how they decide under uncertainty, where the boundaries of safe practice sit. We bring together expertise from psychology, mental health, philosophy and technology to build the high-level cognitive models that adherence depends on.
Reasoning first, expressed as atomic behavioural loops.
We start from the reasoning, not the output, and express it as atomic behavioural patterns and loops: small, well-bounded units of perceiving, deciding, acting and checking, each carrying only the context it needs. Bounded units keep the conditions that should govern behaviour from being crowded out by ordinary content, and they compose into workflows that hold their adherence as complexity grows.
Adherence and context architecture.
We study how AI systems can remain faithful to the conditions that define appropriate behaviour. We treat broad alignment as a larger, emergent challenge — and focus on the practical layer where product failures can be observed, measured and improved: context design, constraint preservation, role-specific reasoning, staged workflows and verification.
Two demanding domains, one shared problem.
Sales intelligence
We apply adherence research to data qualification, account research, ICP pre-checks, role-specific reasoning, copy generation, verification and scheduling. The goal is to help AI behave more like a bounded commercial operator: clear about the task, the buyer context, the evidence base and the limits of its role.
Therapeutic & reflective technology
We apply adherence research to high-sensitivity reflective contexts, where AI must preserve safety constraints, avoid role confusion, support human judgement and collaborate within clinical or quasi-clinical boundaries. The goal is to understand how AI can aid reflection while maintaining appropriate limits.
From context as content to context as governing conditions.
A narrow framework for a recurring failure mode in large language model applications: durable constraints are often stored and processed in the same context substrate as transient task content. System instructions, role boundaries, safety criteria, domain rules and user context then compete with ordinary conversational material inside a finite attention window.
The central claim is that adherence requires systems to distinguish between content that should be reasoned about and contextual conditions that should govern how reasoning proceeds — separating, staging and verifying different kinds of context so governing conditions are less easily displaced by transient material.
Type A — content competes with other content.
Type B — content competes with a constraint, value-prior, role boundary or safety condition. Type B is where adherence breaks down, and where bounded-context architecture helps most.
Scope, deliberately bounded: broad AI alignment remains outside the immediate claims. Context architecture is one safety layer within a wider strategy. Commercial and therapeutic systems remain separate in data, users and workflows.