Nothing enters the ERP without a human signing it
What separates iLEAN from an "autonomous" AI agent is that nothing captured enters the central system without a human seeing it first. In a Tier-2 stamper that becomes an industrial tablet in every booth or line station: two taps — check (✓) to validate, pencil (✏️) to correct — between what the AI captures and the ERP master.
The fear is not AI — it is AI writing on its own and contaminating the ERP and the IATF history.
In a Tier-2 automotive stamping plant, the ERP master and the IATF 16949 history hold the traceability of every heat, every mill certificate and every lot of parts shipped to the seating and suspension Tier-1s. Any figure that enters wrong — a misidentified heat, a mistranscribed first-off, a badly recorded specification change — propagates downstream until somebody catches it. That is why the cultural blocker here is twofold:
- The fear of AI writing data on its own — it is the biggest cultural blocker to digitizing the plant. If a system writes into the master by itself and a figure later turns out to be wrong, the cost is not correcting a field: it is reopening the IATF evidence for everything that depended on it in front of the next auditor of each OEM's requirements. It is the committee's first question: who has seen this before it existed in the ERP?
- But the operator will not type between strokes either — at the press, between one cycle and the next, nobody is going to fill in forms or open menus. If the alternative to AI is transcribing by hand, the data arrives late, incomplete or not at all — and out of mistrust many captures are never recorded in the first place.
Full automation does not pass the plant's cultural filter. Going back to paper is no longer an option. The answer is not choosing an extreme — it is giving the operator a fast, human gate between the AI and the central system.
Connect in early-human-verification mode — the person before the master, not after.
Everything iLEAN does rests on three safety rings: Connect carries the data as it arrives (photo, voice, parsed email, panel reading), Agents interprets it and decides what action to propose, and the person validates before it crosses into the ERP. That separation is the backbone of the whole iLEAN stamping matrix — and this piece is where the third ring becomes visible and tangible: an industrial tablet in every press booth or line station, with a two-tap gesture on every capture.
An industrial tablet in every press booth. A two-second visual summary. Two taps — check (✓) to validate, pencil (✏️) to correct — before the data crosses into the ERP.
How early human verification works in a Tier-2 automotive stamping plant:
- An industrial tablet in every booth or line station — a UI designed for a two-second validation, not for data entry. There is no keyboard and no nested menus: there is a visual summary and a two-tap decision.
- A visual summary (heat, lot, event) — whether the data came from a photo, from voice, from a parsed email or from a panel reading, Connect normalizes it into the same simple format before showing it to the operator.
- Check (✓) to validate, pencil (✏️) to correct — if the capture came through clean, one tap and they carry on; if something does not add up, the correction is minimal and the original is filed alongside it.
- Latency between capture and a clean master: seconds — validated data reaches the ERP almost at the same time it was generated, not at the end of the shift.
- No validation, no master — if the operator does not validate, the data is held and contaminates nothing downstream. The press carries on; the data waits.
- Verification designed for validating, not typing — the operator keeps their booth rhythm and their judgment, and exercises it on real information from their own press rather than on what gets remembered at shift end.
Unvalidated data (or data never captured out of mistrust) vs. data validated in seconds
| Aspect | Without early human verification | With iLEAN Connect + tablet |
|---|---|---|
| Capture by photo/voice/email/panel | Discarded out of mistrust, or enters straight through unreviewed | Always passes through a visual summary and two taps before touching the master |
| Operator recording | Typing between press strokes — which nobody is going to do | Check (✓) or pencil (✏️) on the tablet, no typing |
| Wrong or doubtful data | Contaminates the ERP and the IATF history and drags evidence downstream | Held — it does not cross into the master until somebody validates it |
| Time between capture and a clean master | Hours or shifts, if it ever gets recorded | Seconds |
| IATF evidence at audit | With gaps — uncaptured data or no trace of who validated | No gaps — every figure validated by a human, with who, when and what changed |
| The plant's stance on AI | Blocked — "we do not know what is going in by itself" | A clean master: every figure has passed through a human before existing |
The value of this piece — enabling, not a payback in months.
This case is deliberately not presented with a payback figure. Its value is structural: it is what lets everything else be signed off.
- It is the cross-cutting enabling piece: every Connect, Edge or Agents case in your stamping plant depends on this ring existing so the data they capture can enter the ERP and the IATF history without fear.
- It does not replace work — it replaces doubt. The cost it avoids is not an hour of operator time: it is the paralysis of "we are not going to let an AI write into our IATF master on its own".
- It is also the piece of cultural change the line requires: the operator does not inherit forms to type between strokes, they inherit a two-tap tablet — and they keep their booth rhythm and their judgment.
- The indicators that can be measured from the first pilot — as an estimate to be validated in each plant: the percentage of captures validated in under five seconds, and the percentage of held data that turned out to be real corrections — the latter measures whether capture is well calibrated, not whether "the AI fails".
- We do not put forward a payback in months here — that would be inventing a number only your plant's assessment can give. What we can state is that without this piece, the rest of the IRIS roadmap in your stamping plant stays a pilot and never scales.
And the fair question from the committee
"Why not automate fully, if we already have AI that reads well?" — because reading well is not the same as deciding well, and deciding well is not the same as being authorized to write into the ERP master with nobody reviewing it. iLEAN separates the three functions deliberately: Connect carries the data as it arrives, Agents interprets it and proposes the action, and the person validates before it crosses into the master. Those are the three safety rings, and this piece is where the third becomes physical: a tablet, a two-second summary, a check or a pencil. Hallucination is a problem of free generation, not of anchored tasks — in tasks where the AI merely recontextualizes a figure (reading a panel, transcribing a label, parsing an email), the best models brought the error below 1.5% [1]. And even so, in a press booth what is critical is not recorded on its own: that is why the tablet exists — and why the last word still belongs to the operator.
[1] OpenAI paper "Why Language Models Hallucinate", 2025 — on the reliability of AI in anchored tasks.
What people ask about the early human verification tablet
Why is early human verification the key cultural blocker to digitizing a stamping plant?
Because in a Tier-2 stamper the ERP master and the IATF 16949 history hold the traceability of every heat, every mill certificate and every lot of parts shipped to the seating and suspension Tier-1s. The fear of AI writing data on its own and contaminating that master is the biggest cultural blocker to digitization — it is the committee's first question: who has seen this before it existed in the ERP? A single wrongly written figure does not cost one field: it costs reopening the IATF evidence for everything that depended on it at the next audit of each OEM's requirements. Early human verification defuses that fear by putting the person before the master, not after: iLEAN separates Connect (carries the data as it arrives), Agents (interprets and proposes the action) and the person (validates in two taps before the data crosses into the central system). Without that third ring made visible, no other piece of the matrix gets signed off — which is why it is the enabling piece.
How is this different from an "autonomous" AI agent that writes straight into the ERP?
In that an autonomous agent decides and writes with nobody reviewing it, and that turns a one-off error into a systemic one: in stamping, a badly recorded heat or lot propagates downstream — first-off, CMM, shipment to the Tier-1 — until somebody catches it, and by then correcting it is no longer editing a field, it is redoing the IATF evidence. iLEAN deliberately does not automate fully: Connect carries the data, Agents interprets it and proposes the action, and the person validates before it crosses into the master. The AI proposes, the operator disposes. The tablet UI is designed for validating, not for data entry — it is the opposite of an agent that skips the person. Hallucination is a problem of free generation, not of anchored tasks: when the AI merely recontextualizes a figure (reading a panel, transcribing a label, parsing an email), the error is low — but even so, in a press booth what is critical is not recorded on its own.
What do the "two taps" really look like without the operator typing between strokes?
Two taps means exactly that: every capture — photo, voice, parsed email — appears on the tablet as a two-second visual summary with two possible gestures, check (✓) to validate or pencil (✏️) to correct. The UI is designed for a two-second validation, not for data entry: there is no keyboard in the normal flow, no nested menus, no mandatory fields to fill. The operator does not want to type between press strokes, and they do not have to: if the capture came through clean — which is the usual case — they look, tap the check and carry on with their booth. Only when something does not add up does the pencil come in, opening a minimal correction on the specific field rather than a whole form. It is measurable from the first pilot: the percentage of validations resolved in under five seconds is one of the startup indicators, as an estimate to be validated in each plant.
What happens when the operator corrects a figure — is the AI's original capture lost?
No. When the operator taps the pencil (✏️) and corrects, the original figure Connect captured is filed alongside the correction, not overwritten — the ERP receives the validated figure and the history keeps both, with who validated, when and what changed. This matters for two reasons. First, stamping IATF evidence — from the heat to the lot shipped to the Tier-1 — requires being able to demonstrate what happened, not only what was corrected: an 8D or a PFMEA rests on that trail. Second, it makes it possible to detect with evidence whether capture fails repeatedly at one specific point — a hard-to-read coil label, a press panel with glare — and adjust the capture, rather than arguing over impressions. The operator's correction is not a patch: it is the signal that calibrates the system and keeps the master clean and the IATF evidence free of gaps.
Why does this tablet enable the other pieces of the iLEAN stamping matrix?
Because it is the cross-cutting enabling piece: every Connect, Edge or Agents case in your stamping plant — JIDOKA AI on the press, assisted SMED, first-off control, CMM reading — depends on this ring existing so the data they capture can enter the ERP and the IATF history without fear. Without early human verification, the other eleven pieces of the matrix do not get signed off, because nobody authorizes their data to enter the master: they stay pilots and never scale. It does not replace work — it replaces doubt: the cost it avoids is not an hour of operator time, it is the paralysis of "we are not going to let an AI write into our IATF master on its own". Its value is strategic, not a payback in euros per incident avoided: we do not put forward months of return because that would be inventing a number only your plant's assessment can give. What we can state is that without this piece, the rest of the IRIS roadmap in your stamping plant never gets started.
Two operator taps, a clean master every time — how many press booths do you have? We will quote tablets plus UI for your plant.
We work on your plant's real data, not ours. Assessment with no commitment.
See the tablet in the press booth ‹ See all 12 metal stamping cases See automotive