Skip to content
Docs
Governance Conformance Brief

Smartflow × EU Digital Omnibus on AI

A requirement-by-requirement map of the amended EU AI Act to Smartflow's enforcement plane and EU AI Act Conformity Console, with conservative verdicts and the gap to close. Companion to the IMDA mapping.

Source: PE-CONS 30/26 · Reg. (EU) 2026/… · Digital Omnibus on AI · agreed 18 Jun 2026 Amends: (EU) 2024/1689 (AI Act), 2018/1139 (EASA), 2023/1230 (Machinery) Procedure: 2025/0359(COD) Status: v1.7.55 · Conformity Console live · ID.me human-in-the-loop live (cluster1)

What the Omnibus actually changes

The Digital Omnibus is a simplification package, not a rollback. It lightens process and paperwork while leaving the substantive runtime obligations intact — and in two places it makes them stricter and more enforceable:

Why this is a tailwind for Smartflow

The Omnibus removes paperwork but doubles down on the two things Smartflow is: a deterministic runtime enforcement plane between the agent/model and the systems it touches, and a tamper-evident, machine-readable evidence layer a supervisor can inspect on demand.

"… reasonable and adequate technical safety measures … including runtime prompt guardrails, content classification and filtering mechanisms, usage restrictions, abuse detection mechanisms, and notice and action mechanisms." — Recital 12, p.12
"… order operators to provide access to … their AI systems … impose an obligation on an operator to retain all data and documents deemed to be necessary to assess … compliance." — Art 75a(6), p.77

Smartflow already ships the EU AI Act Conformity Console (src/conformity_api.rs, /api/conformity/* + /conformity dashboard) surfacing regulator-facing evidence for Articles 9, 10, 11/Annex IV, 12, 13, 14, 15, 17, 26 and 79 — the exact articles this Omnibus restructures.

EXISTS enforced / evidenced on the request path today PARTIAL present but opt-in / shadow / partial coverage ABSENT not a gateway concern / product-side

What changed since the original AI Act (Reg. (EU) 2024/1689)

Delta view: the substantive runtime obligations survive; the changes are scope-tightening, burden relief, two new prohibitions, and a much stronger central enforcement + evidence regime.

ProvisionOriginal AI Act (2024/1689)Digital Omnibus change (PE-CONS 30/26)Smartflow angle
Art 5 — Prohibited practicesFixed list (manipulation, social scoring, untargeted scraping, etc.)Adds (ba) non-consensual intimate material & (bb) CSAM; providers must hold "reasonable and adequate technical safety measures" where misuse is foreseeable (Art 5(1a)); Recital 12 names runtime guardrails, content filtering, usage limits, abuse detectionDirectly our enforcement plane — see Block A
Art 4 — AI literacyBinding duty to "ensure a sufficient level of AI literacy"Softened to "take measures to support" AI-literacy developmentLower friction for customers; not a gateway feature
Art 4a (new) — Bias-detection dataArt 10(5): special-category processing limited to providers of high-risk systemsExtended to deployers & other AI/models, under strict access-control, documentation & deletion safeguardsAccess logging + information barriers evidence the safeguards
Art 3(14), 6(1a–1c) — Safety component / high-risk scopeBroad "safety component" definitionClarified & narrowed; non-safety functions (assistance, optimisation, convenience, QC) don't trigger high-riskFewer systems in scope, but those in scope still need runtime control
Art 2(13) (new) — Equivalence limitationArt 9–15 / 17–25 apply in fullCommission may limit these where Section A product law gives equivalent protection (delegated acts by 2 Aug 2027)Console maps to whichever set remains in force per product
Art 11 / 17 / 63 — Docs & QMSFull Annex IV documentation & QMSSimplified Annex IV & proportionate QMS for SMEs/SMCs; new SME/SMC definitions (Art 3(14a/14b))Auto-generated evidence lowers the bar further
Art 28 / 29 / 43 — Conformity assessmentSeparate designations per legislationSingle application + unified assessment; new Annex XIV code taxonomy (AIP/AIB/AIH)Process change; no gateway impact
Art 42(3) (new) — Cybersecurity presumptionArt 15 cybersecurity assessed independentlyHigh-risk AI in scope of the Cyber Resilience Act (2024/2847) deemed to meet Art 15 cybersecurityComplements our robustness/abuse controls
Annex I / MachineryMachinery (2023/1230) in Section AMoved to Section B (sectoral approach); delegated acts by 2 Aug 2028EASA & machinery now sector-routed
Art 50(7) — Synthetic-content markingCodes of practice encouragedCommission may set common rules if codes are inadequate (implementing act)Marking is product/provider-side (watch item)
Art 72(3) — Post-market monitoringHarmonised plan template mandatedTemplate empowerment removed; flexible plan, guidance by 2 Sep 2027Our monitoring/analytics feed the plan
Art 75 + 75a–75d (new) — Central enforcementMainly national market-surveillance authoritiesAI Office exclusively competent for GPAI-based systems & VLOPs; direct investigation/inspection powers, order access & retention, fines + periodic penalties up to 5% daily turnoverTamper-evident chain = the retained record on demand
Art 77 — Fundamental-rights accessAuthorities may request informationMust be provided in machine-readable format by electronic meansExactly the Conformity Console output
Entry into applicationStaggered to 2 Aug 2027Omnibus enters into force urgently (3rd day after publication); key delegated acts 2 Aug 2027, machinery 2 Aug 2028Buyers need a compliance answer now

Block A · New & sharpened
Article 5 prohibitions — the runtime-safeguard expectation

Omnibus requirement (page)Smartflow capability / evidenceVerdict
Runtime prompt guardrails against prohibited generation (Recital 12, p.12; Art 5(1a))Maestro semantic policy corpus + request-side guardrail graph; Shield destructive-prompt rules — request-path enforcementEXISTS
Content classification & filtering of inputs/outputs (Recital 12)compliance_detector — PII / toxic / regulated-content classification at ingest and egress (MCP_COMPLIANCE_SCAN_ENABLED); perimeter classification on egress (src/perimeter.rs)EXISTS
Usage restrictions & abuse detection (Recital 12)Per-key/route allowlists, budgets, rate limiter + adaptive rate limiter; Retrospector multi-turn abuse passesEXISTS
Notice-and-action on observed/reported misuse + corrective action on circumvention (Recital 12, p.12)Blocked-action sealing into the VAS hash chain + alerting fan-out; Shield approval/override queue closes the loopPARTIAL
Consent collection for lawful intimate-material generation; synthetic-content marking (Art 5(1a); Art 50)Identity/consent attestation possible via ID.me step-up, but consent capture & C2PA-style content watermarking are product/provider-sideABSENT

Block B · Preserved high-risk duties
Articles 9–17 — risk, data, docs, logging, transparency, oversight, robustness

AI Act article (as amended)Smartflow capability / evidence (Conformity Console)Verdict
Art 9 — Risk-management systemGuardrail policy graph, Maestro corpus, Shield tiered destructive-op rules; GET /api/conformity/articles/art_9EXISTS
Art 10 + new Art 4a — Data governance & bias-detection safeguardscompliance_detector data-class scopes; information-barrier (LDAP-group × content-label); AIDA signed data-class scopes. Art 4a access-control/documentation/deletion safeguards partly covered by access loggingPARTIAL
Art 11 / Annex IV — Technical documentation (simplified for SMEs/SMCs)Examination suite + AI inventory + signed AI-BOM; JSON export; GET /api/examinations, /api/ai_inventoryPARTIAL
Art 12 / Art 19 — Automatic record-keeping over lifecycleVAS per-request log (input/output/model/identity/decisions) + audit_chain HMAC-SHA256 prev-hash chain; independently re-runnable audit-verifier; POST /api/admin/audit/verify_chainEXISTS
Art 13 — Transparency / information to usersMaestro response banners; x-smartflow-group-banner, x-shield-warn headers for downstream UIPARTIAL
Art 14 — Human oversight (effective oversight by natural persons)Shield High-severity → human approval queue; live ID.me human-in-the-loop step-up (IAL2/AAL2 + liveness) holds high-risk LLM/MCP calls until a verified human approves; Phase-4 oversight analytics (override rate / approval latency / outlier reviewers)EXISTS
Art 15 + new Art 42(3) — Accuracy, robustness, cybersecurity (CRA presumption)AI analyzer (hallucination confidence), A/B regression detection, alerting; Shield blocks destructive prompts + tool calls; rate limiter / circuit breakerPARTIAL
Art 17 / Art 63 — Quality-management system (proportionate for SMEs/SMCs)Retrospector 5-pass async QMS program; examination framework reports; alerting pipelines; GET /api/retrospector/passesPARTIAL

Block C · Deployer, monitoring & incidents
Articles 26, 72, 73/79

AI Act article (as amended)Smartflow capability / evidenceVerdict
Art 26 — Deployer obligations (human oversight, monitor, keep logs)ID.me-attested deployer natural-person identity; every request stamped with user_id/groups/department; tamper-evident logs of deployer actions; GET /api/audit/logs?event_type=verified_identityEXISTS
Art 72 — Post-market monitoring (flexible plan)VAS analytics + perimeter egress monitoring + Retrospector nightly org report; no formal post-market-monitoring-plan export yetPARTIAL
Art 73 / 79 — Serious-incident reporting (incl. to AI Office, Art 75(1a))Critical shield/nipr audit events + severity≥High Retrospector findings + webhook fan-out; pre-formatted incident envelope at GET /api/conformity/incidentsEXISTS

Block D · The new enforcement regime
Articles 75a–75d & 77 — evidence on demand

Omnibus power (page)Smartflow capability / evidenceVerdict
AI Office may order access to the AI system & retain all data/documents (Art 75a(6), p.77)Conformity Console read-only API surfaces live evidence per article; tamper-evident audit chain provides the retained record on demandEXISTS
Fundamental-rights bodies may demand evidence in machine-readable format (Art 77(1), p.84)/api/conformity/* returns structured JSON evidence + honest gap list per article; dashboard at /conformityEXISTS
Inspections / proof of tamper-free records over timeIndependently rebuildable audit-verifier (regulator-runnable). Gap: WORM retention (S3 Object Lock / Azure immutable) + RFC 3161 trusted timestamps on roadmap (B5/B6)PARTIAL
Ability to take a system offline / limit blast radius during enforcementGateway-wide emergency stop (off / read-only / halt) across proxy + MCP + A2A, audit-chained (src/emergency_stop.rs, POST /api/admin/emergency-stop)EXISTS

Summary verdict

#Omnibus themeVerdictOne-line position
1Art 5 runtime safeguards (NCII/CSAM)EXISTSGuardrails, content filtering, usage limits, abuse detection on the request path; consent/marking is product-side
2High-risk duties Art 9–17PARTIALConsole evidences all of them; Art 9/12/14 fully on hot path, rest partial/opt-in
3Human oversight Art 14/26EXISTSLive ID.me human-in-the-loop step-up + oversight analytics — strongest row
4Record-keeping Art 12/19EXISTSTamper-evident chain + regulator-runnable verifier; WORM/RFC-3161 on roadmap
5Incidents Art 73/79EXISTSPre-formatted incident envelope; per-Member-State templates on roadmap
6AI Office evidence-on-demand Art 75a/77EXISTSMachine-readable Conformity Console is exactly what the new powers require
7Burden relief (SME/SMC docs, unified CA)N/AProcess simplification — reduces customer friction to adopt Smartflow, not a gateway feature

Net assessment

The Digital Omnibus is net-positive for APERION. It strips out paperwork that never differentiated us, while sharpening the two demands only a runtime layer can satisfy: (1) Article 5 now explicitly expects "runtime prompt guardrails, content filtering, usage restrictions and abuse detection" — Smartflow's core — and (2) the AI Office's new powers to compel access, retention and machine-readable evidence map one-to-one onto our tamper-evident audit chain and the /api/conformity/* console. The headline runtime story — holding a high-risk action until a government-grade verified human (ID.me IAL2/AAL2) approves it, now live on cluster1 — is a direct, demonstrable answer to Article 14 human-oversight effectiveness. Honest gaps remain on the evidentiary hardening track (WORM retention, RFC 3161 timestamps) and on provider-side concerns the gateway does not own (training-data lineage, synthetic-content marking, consent capture). Closing the hardening track yields legal-grade, on-demand conformance evidence for an EU-published, freshly-simplified regime — a strong differentiator for regulated buyers and SMEs/SMCs the Omnibus is explicitly courting.