01
LoadLine
LoadLine is a standing risk instrument for organizations adopting AI: ARQ's assessors, working with your team on a weekly rhythm, examine the AI pathways you declare and rank them against the tolerance your leadership defines, by business impact and by how far each failure chain has progressed, so the decisions your leadership faces arrive in order of consequence.
It is a capability, people plus instrument, not a software product and not a one-time assessment. The instrument stays: as your team enters new evidence, the record updates, so the picture your leaders examine is the current one.
Where ARQ Lab fits: LoadLine™ is ARQ Lab's instrument. Independent by structure: we don't build your AI, so we're not grading our own work.
02
LoadLine Readiness Index (LRI™)
The LoadLine Readiness Index (LRI™) is the executive-level reading LoadLine produces: a ranked, evidence-gated view of an organization's declared AI exposure, built for the leaders above the CISO, the CEO, CFO and CRO, rather than for practitioners.
Evidence-gated means every reading answers "can the organization prove it?", not self-attested questionnaire scores.
Where ARQ Lab fits: the LRI™ is what a LoadLine engagement hands to the leadership team.
03
Failure Pathway Intelligence (FPI™)
Failure Pathway Intelligence (FPI™) is the discipline of mapping the hidden pathways by which small governance inconsistencies compound into material organizational failure.
Its core insight: AI failures don't occur as isolated events. They propagate across policy, operations, technology, governance, vendors and people. FPI makes those propagation paths visible before they complete.
Where ARQ Lab fits: FPI is the discipline LoadLine applies.
04
The Detection Zone
The Detection Zone is the span of an AI failure pathway where a developing failure is still detectable and preventable, before an automated decision executes and becomes business impact.
The pathway runs: weak identity, AI manipulation, incorrect trusted output, missed detection, automated decision executes, business impact. Everything to the right of the Detection Zone is expensive; everything inside it is still preventable.
Where ARQ Lab fits: ARQ Lab works inside the Detection Zone, before impact.
05
Never-paste list
A never-paste list is a short, written list of the things that never go into a consumer AI tool, no matter how convenient, kept where you work, so the decision is already made before the moment of temptation.
A starter list: client names, anything about your employees, anything you'd call a trade secret.
Where ARQ Lab fits: the never-paste list is a practice ARQ Lab teaches in Adult Supervision.
06
Adult Supervision
Adult Supervision is ARQ Lab's education newsletter on supervising AI at work, with a podcast to follow: short lessons for leaders whose AI is brilliant, tireless and unsupervised.
Written by Dave Tyson as a fellow traveler, not a lecturer. Each lesson is under 400 words with a ten-minute exercise.
Subscribe on LinkedIn →
07
Bivalent
Bivalent is ARQ Lab's control architecture for autonomous AI agents, built on one principle: the actor doesn't hold the switch. Agents receive a clearance at the point of action, and every decision leaves a receipt you can produce on demand.
Observation tools act as the commentator, reporting what went wrong afterward. Bivalent is built to be the referee. Patent pending.
Where ARQ Lab fits: Bivalent is ARQ Lab's instrument for holding the line.
08
Digital employee
A digital employee is an AI worker built to a specific operation with three things defined before it starts: the job, the limits and the supervisor.
Illustrative roles are illustrations, not case studies. Every digital employee is scoped in a discovery session, where a leadership team maps its workflows against a disciplined test of what an AI worker can genuinely carry.
Where ARQ Lab fits: ARQ Lab builds digital employees under the same governance it sells.
