Human in the loop
Every decision an agent refused to take alone
An escalation is not a failure. It is the system working: the agent reached the edge of its authority, stopped, wrote down what it recommends and why, and handed a named human a decision with the options already costed. What matters is how long the human takes and whether the work went back to the agent afterwards.
Health
What has been handed to a person
Escalations by tier and status, with how long the oldest have been waiting.
Open now
24
Acknowledged
11
Resolved
10
Returned to agent
11
human decided, agent resumed
Past SLA
35
the human is the bottleneck
Units behind them
6,548
blast radius if nothing moves
Territory & Quota Planning: SL1B-1006 held at draft delta
Territory at draft delta valued at $1.9m stopped on a envelope breach. The lane cleared 53 percent of 56 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1h 2m
SLA 18h 3m
What the agent recommends
White Space Agent recommends rejecting and returning to the requester with the gap named.
Confidence 95 percent. The comparable population is 186 units; 91 percent of them cleared on the same pattern. What separates this one is a control that has failed twice this quarter.
Options put in front of the human, already costed
Account Planning: SL2C-1003 held at assemble context
Account plan at assemble context valued at $16k stopped on a envelope breach. The lane cleared 72 percent of 4079 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
14h 13m
SLA 44m · breached
What the agent recommends
Account Research Agent recommends rejecting and returning to the requester with the gap named.
Confidence 88 percent. The comparable population is 398 units; 87 percent of them cleared on the same pattern. What separates this one is a jurisdiction the policy version does not cover.
Options put in front of the human, already costed
Brand & Positioning: MK1C-1005 held at draft readout
Brand asset at draft readout valued at $869k stopped on a segregation of duties. The lane cleared 44 percent of 34 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 5h
SLA 18h 44m · breached
What the agent recommends
Narrative Drafting Agent recommends approving a partial and holding the remainder.
Confidence 92 percent. The comparable population is 187 units; 78 percent of them cleared on the same pattern. What separates this one is a counterparty with no prior history.
Options put in front of the human, already costed
Product Marketing: MK7B-1003 held at model options
Launch at model options valued at $2.2m stopped on a confidence floor. The lane cleared 37 percent of 4 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 0h
SLA 3d 0h
What the agent recommends
Persona Research Agent recommends approving a partial and holding the remainder.
Confidence 88 percent. The comparable population is 433 units; 85 percent of them cleared on the same pattern. What separates this one is a jurisdiction the policy version does not cover.
Options put in front of the human, already costed
Approved with a compensating control after 180 minutes. Rationale and clause recorded on the unit.
Commercial Advisory: LG2B-1004 held at frame
Advisory matter at frame valued at $17.2m stopped on a confidence floor. The lane cleared 25 percent of 2 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 5h
SLA 5d 17h
What the agent recommends
Question Triage Agent recommends approving with the compensating control attached.
Confidence 95 percent. The comparable population is 292 units; 67 percent of them cleared on the same pattern. What separates this one is a control that has failed twice this quarter.
Options put in front of the human, already costed
Returned with a corrected policy version. The lane re-ran it without a human.
Compliance & Ethics: LG4B-1004 held at score risk
Compliance case at score risk valued at $63k stopped on a novel pattern. The lane cleared 75 percent of 6427 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
10h 4m
SLA 24m · breached
What the agent recommends
Gifts & Hospitality Agent recommends pausing the lane until the upstream data defect is fixed.
Confidence 86 percent. The comparable population is 265 units; 89 percent of them cleared on the same pattern. What separates this one is a jurisdiction the policy version does not cover.
Options put in front of the human, already costed
IP Portfolio: LG6C-1001 held at challenge
IP matter at challenge valued at $40.9m stopped on a policy clause engaged. The lane cleared 41 percent of 1 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
22h 4m
SLA 3d 22h
What the agent recommends
Infringement Watch Agent recommends approving a partial and holding the remainder.
Confidence 89 percent. The comparable population is 451 units; 72 percent of them cleared on the same pattern. What separates this one is a value above the envelope.
Options put in front of the human, already costed
Privacy & Data Protection: LG9C-1026 held at regulator notify
Privacy case at regulator notify valued at $81k stopped on a confidence floor. The lane cleared 65 percent of 1669 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 0h
SLA 5h 41m · breached
What the agent recommends
Data Protection Officer recommends pausing the lane until the upstream data defect is fixed.
Confidence 55 percent. The comparable population is 820 units; 76 percent of them cleared on the same pattern. What separates this one is a control that has failed twice this quarter.
Options put in front of the human, already costed
Category Strategy: PR1B-1000 held at draft position
Category plan at draft position valued at $38.7m stopped on a segregation of duties. The lane cleared 37 percent of 4 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
12h 40m
SLA 3d 19h
What the agent recommends
Should-cost Modeling Agent recommends pausing the lane until the upstream data defect is fixed.
Confidence 93 percent. The comparable population is 619 units; 91 percent of them cleared on the same pattern. What separates this one is a counterparty with no prior history.
Options put in front of the human, already costed
Returned with a corrected policy version. The lane re-ran it without a human.
Negotiation & Award: PR4A-1010 held at award
Negotiation at award valued at $35.9m stopped on a confidence floor. The lane cleared 28 percent of 3 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
21h 31m
SLA 6d 7h
What the agent recommends
VP Sourcing recommends pausing the lane until the upstream data defect is fixed.
Confidence 54 percent. The comparable population is 289 units; 65 percent of them cleared on the same pattern. What separates this one is a value above the envelope.
Options put in front of the human, already costed
Returned with a corrected policy version. The lane re-ran it without a human.
Payment Interlock: PR10C-1004 held at treasury review
Payment run at treasury review valued at $712k stopped on a segregation of duties. The lane cleared 55 percent of 56 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
13h 20m
SLA 2d 10h
What the agent recommends
Director, Payment Operations recommends rejecting and returning to the requester with the gap named.
Confidence 57 percent. The comparable population is 375 units; 68 percent of them cleared on the same pattern. What separates this one is a jurisdiction the policy version does not cover.
Options put in front of the human, already costed
Portfolio & Intake: EN1A-1013 held at commit to sprint
Request at commit to sprint valued at $58k stopped on a break condition fired. The lane cleared 72 percent of 4388 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 5h
SLA 9h 20m · breached
What the agent recommends
VP Engineering Operations recommends pausing the lane until the upstream data defect is fixed.
Confidence 47 percent. The comparable population is 390 units; 76 percent of them cleared on the same pattern. What separates this one is a value above the envelope.
Options put in front of the human, already costed
Architecture & Design: EN2A-1001 held at model options
Design decision at model options valued at $16.9m stopped on a policy clause engaged. The lane cleared 28 percent of 5 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
11h 25m
SLA 3d 9h
What the agent recommends
Pattern Compliance Agent recommends approving with the compensating control attached.
Confidence 92 percent. The comparable population is 285 units; 63 percent of them cleared on the same pattern. What separates this one is a jurisdiction the policy version does not cover.
Options put in front of the human, already costed
Innovation Portfolio: RD2A-1004 held at draft recommendation
Program at draft recommendation valued at $88k stopped on a confidence floor. The lane cleared 67 percent of 5106 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
7h 4m
SLA 4h 2m · breached
What the agent recommends
Portfolio Balance Agent recommends approving a partial and holding the remainder.
Confidence 85 percent. The comparable population is 216 units; 71 percent of them cleared on the same pattern. What separates this one is a conflicting record in the master data.
Options put in front of the human, already costed
Prototype & Pilot: RD6B-1052 held at scale decision
Prototype at scale decision valued at $84k stopped on a envelope breach. The lane cleared 77 percent of 475 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
4h 28m
SLA 2h 15m · breached
What the agent recommends
Director, Product Development recommends pausing the lane until the upstream data defect is fixed.
Confidence 42 percent. The comparable population is 612 units; 70 percent of them cleared on the same pattern. What separates this one is a control that has failed twice this quarter.
Options put in front of the human, already costed
Approved with a compensating control after 133 minutes. Rationale and clause recorded on the unit.
Regulatory Science: RD8C-1005 held at non-conformance escalate
Submission at non-conformance escalate valued at $62k stopped on a policy clause engaged. The lane cleared 64 percent of 7479 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
5h 11m
SLA 2h 50m · breached
What the agent recommends
VP Regulatory Science recommends approving a partial and holding the remainder.
Confidence 59 percent. The comparable population is 536 units; 78 percent of them cleared on the same pattern. What separates this one is a conflicting record in the master data.
Options put in front of the human, already costed
FP&A & Planning: FI1B-1005 held at leader review
Analysis at leader review valued at $286k stopped on a policy clause engaged. The lane cleared 54 percent of 84 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
8h 28m
SLA 1d 23h
What the agent recommends
VP Financial Planning & Analysis recommends pausing the lane until the upstream data defect is fixed.
Confidence 70 percent. The comparable population is 502 units; 80 percent of them cleared on the same pattern. What separates this one is a conflicting record in the master data.
Options put in front of the human, already costed
Returned with a corrected policy version. The lane re-ran it without a human.
Controllership & Policy: FI5C-1001 held at controller approve
Technical position at controller approve valued at $56k stopped on a novel pattern. The lane cleared 60 percent of 7763 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 4h
SLA 10h 29m · breached
What the agent recommends
Chief Accounting Officer recommends approving a partial and holding the remainder.
Confidence 49 percent. The comparable population is 473 units; 65 percent of them cleared on the same pattern. What separates this one is a conflicting record in the master data.
Options put in front of the human, already costed
Treasury & Cash: FI6B-1002 held at model hedge
Cash action at model hedge valued at $73k stopped on a break condition fired. The lane cleared 77 percent of 3944 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
7h 44m
SLA 2h 59m · breached
What the agent recommends
Liquidity Forecast Agent recommends approving with the compensating control attached.
Confidence 93 percent. The comparable population is 645 units; 77 percent of them cleared on the same pattern. What separates this one is a counterparty with no prior history.
Options put in front of the human, already costed
Tax: FI7B-1000 held at file
Tax filing at file valued at $53k stopped on a confidence floor. The lane cleared 68 percent of 7327 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 4h
SLA 1h 45m · breached
What the agent recommends
Provision Agent recommends approving a partial and holding the remainder.
Confidence 85 percent. The comparable population is 881 units; 91 percent of them cleared on the same pattern. What separates this one is a value above the envelope.
Options put in front of the human, already costed
Strategic Workforce Planning: HR1C-1006 held at draft position
Plan scenario at draft position valued at $16.4m stopped on a confidence floor. The lane cleared 30 percent of 1 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
5h 30m
SLA 2d 9h
What the agent recommends
Build-Buy-Borrow Agent recommends approving a partial and holding the remainder.
Confidence 90 percent. The comparable population is 781 units; 91 percent of them cleared on the same pattern. What separates this one is a conflicting record in the master data.
Options put in front of the human, already costed
Total Rewards: HR8A-1028 held at set budget
Comp action at set budget valued at $47k stopped on a segregation of duties. The lane cleared 68 percent of 5677 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 7h
SLA 9h 8m · breached
What the agent recommends
VP Total Rewards recommends approving with the compensating control attached.
Confidence 53 percent. The comparable population is 109 units; 95 percent of them cleared on the same pattern. What separates this one is a counterparty with no prior history.
Options put in front of the human, already costed
Approved with a compensating control after 54 minutes. Rationale and clause recorded on the unit.
Executive Support: AD4B-1054 held at approve
Executive request at approve valued at $65k stopped on a policy clause engaged. The lane cleared 75 percent of 4287 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
22h 33m
SLA 10h 40m · breached
What the agent recommends
Chief of Staff, Corporate recommends approving with the compensating control attached.
Confidence 58 percent. The comparable population is 311 units; 89 percent of them cleared on the same pattern. What separates this one is a counterparty with no prior history.
Options put in front of the human, already costed
Tier-2 Technical: CS3B-1011 held at package evidence
Escalated case at package evidence valued at $82k stopped on a segregation of duties. The lane cleared 80 percent of 1700 units this week without a human; this one did not clear because the evidence does not carry the decision on its own.
waiting on a person
1d 7h
SLA 5h 17m · breached
What the agent recommends
Log Analysis Agent recommends approving with the compensating control attached.
Confidence 58 percent. The comparable population is 734 units; 77 percent of them cleared on the same pattern. What separates this one is a jurisdiction the policy version does not cover.
Options put in front of the human, already costed
Approved with a compensating control after 101 minutes. Rationale and clause recorded on the unit.
The number worth watching on this page is not how many escalations there are. It is how many are past SLA. A queue of well-formed escalations sitting on a human's desk is the most common way an agentic operating model quietly stops delivering: the agents got faster, the approval layer did not, and the cycle time moved from the process to the person.
Actions
What is waiting on a person
An escalation is an agent saying it will not decide this. Nothing on this list moves until a person answers it.
Pick up an unacknowledged escalation
Raised, nobody has claimed it.
Answer a tier-three escalation
The highest tier on the estate, still unresolved.
Conclude an acknowledged escalation
Somebody owns it. It has not been closed out.
Clear an escalation older than a day
Age measured from the moment the agent stopped.
Operations
What this desk is allowed to start
A surface that only reports is not operable. This is the work this page can set in motion, and the bound it runs into.
Trigger and bound
This desk can start an escalation, route it to a tier, and record the answer it receives. It cannot decide the underlying question, extend an SLA, or close an escalation on the raising agent’s behalf. Returning one to the agent is itself a recorded decision with a named author.
Live observability
What the record shows right now
Where every escalation ever raised currently sits.
Current distribution
56 escalations
Is policy and strategy coming to fruition
Whether the written intent is holding here
10 of 56 escalations have been answered. 35 are still waiting on a person.
Not holding on the record
The design position is that an agent stopping is a healthy outcome, provided a person answers quickly. Measured: 17.9% have been answered, 9 have been open longer than a day, and 22 sit at the highest tier unresolved. The queue is the honest reading of human capacity on this estate — escalation volume is not the problem, answering time is.