EU AI Act · Automotive AI
When an OEM asks whether the approved driving function is still the one in the vehicle, what do you hand them?
Not last year’s approval file. The function running now.
RiskLine records the approved function, watches for change against it, and keeps timestamped proof ready for the supplier-quality review.
Functional safety · Quality · Supplier quality
- 01Function locked
- 02Model retrained
- 03Review pending
- 04Signed off
- 05Function locked
Proof
Walkthrough in production
No automotive systems are seeded yet. Ask for the walkthrough and we will record it with you.
Request this walkthroughWhat your buyer asks
Three questions. Three things you should be able to hand over.
Is it still the approved function?
What you hand over
A record of the intended function, the approved software version, and the operational domain.
Who signs a material change?
What you hand over
A named reviewer before a model or perception change replaces the approved function.
How would you know the envelope moved?
What you hand over
Threshold checks and an incident record after the function is in the field.
Why now
Snapshot assurance does not survive a live system.
Models update faster than approvals
A retrained perception model can ship long before the next formal review cycle catches up.
Suppliers are asked directly
OEM supplier-quality teams ask for the current state of an AI function, not the state at start of production.
Evidence has to carry a date
A change log with a reviewer and a timestamp answers a question a slide deck cannot.
What we verify
Selected obligations → controls → evidence
Sold into European OEMs and Tier 1s. Read by buyers in the UK and US.
| Obligation | Example control | Evidence |
|---|---|---|
| EU AI ActWhere the function is a high-risk safety component | A written record of intended function, approved version and operational domain. | Function record and version lock. |
| Change controlModel or perception-stack change | Review by a named owner before the change replaces the approved function. | Change log with reviewer and a material-change flag. |
| OversightWhere the approved use depends on a driver or operator | The oversight step is named and exceptions are logged. | Oversight status and exception trail. |
| MonitoringOperating envelope in the field | Agreed performance limits for the approved version are checked after deployment. | Threshold checks and an incident record. |
RiskLine does not classify a real vehicle function on this page and makes no claim about ISO 26262 or type approval. Not legal advice. Does not certify compliance.
30-day pilot · optional
If you want to go further, start with one meaningful system.
1–3 AI applications or workflows · 5–15 agreed controls · continuous verification · exception and evidence review · executive findings. No promised legal outcome.
Scope and controls
Agree the systems, the control set, and what “verified” means for each.
Wire verification
Connect classification context and verification checks to the live estate.
Run and exceptions
Continuous checks surface exceptions; owners review and remediate.
Evidence pack
Hand over timestamped findings your buyers can actually read.
Request the walkthrough
Want to see this for Automotive? Tell us who you are.
No pilot needed. Tick the box only if you want to talk about one.
FAQ
Straight answers
Is RiskLine a safety certification?
No. It is an advisory verification layer. It does not replace functional-safety work, type approval or any certification.
Is there an automotive demo?
Not yet. The healthcare demo shows the same chain of system, risk, control and evidence. Request the automotive walkthrough below.
Does it connect to our toolchain?
A pilot agrees which systems and controls are in scope first. Integration follows from that scope.
Another industry?