---
title: 'Operative AI, Safely: Three Risks - SCHILLER'
description: Why direct AI access to ERP data is a documented anti-pattern, what managing
  directors are personally liable for, and how a control layer closes the gap.
url: https://ad-ze.de/en/operative-ai-safe
language: en
---

# Operative AI, safely

## Operative AI, safely.

Three risks every managing director must understand. Before an AI touches your ERP data.

## Connecting an LLM directly to your ERP APIs isn't modern. It's negligent.

An LLM with direct access to your ERP, your email and the open web is a documented security anti-pattern. Data exfiltration happens without a click. Audit trails become incomplete. And you, the managing director, are personally liable. A control layer isn't a convenience. It's the only architecture that meets data protection, industry-specific audit requirements and the EU AI Act at the same time.

What has already happened

## Three documented incidents from 2024 and 2025.

These cases are not hypothetical. They happened. At Microsoft, at Slack, at Replit. With the same products that are being sold to you.

- June 2025 · CVE-2025-32711 

- August 2024 

- July 2025 

Heise also reported in 2025 that Microsoft 365 Copilot did not record file accesses in the audit log when the attacker suppressed the reference link. Microsoft patched silently, without notifying customers. An audit trail with selective gaps is no audit trail.

What you personally risk

## Data protection, EU AI Act, GmbH liability. Three levers, at least one hits you.

- Data protection 

- EU AI Act 

- GmbH liability 

Caveat: not every ERP workflow qualifies as high-risk under the AI Act. Pure order-handling or logistics agents typically don't fall under Annex III. The moment an agent touches HR data, creditworthiness or access to essential services, the high-risk obligations under Annex III apply. High-risk AI in regulated products under Annex I has its own deadlines. Which deadlines apply is set out in the current text of the regulation in the Official Journal of the EU.

What the control layer solves

## The layer makes standard AI traceable.

A control layer between LLM and ERP is not an alternative to Microsoft Copilot or SAP Joule. It is the compliance layer that makes them suitable for regulated industries in the first place.

- 

Tool scoping instead of full access.

The AI assistant gets no direct API keys. It calls defined tools that may do exactly what they are meant to. Nothing more.

- 

PII boundary in both directions.

Recognised personal data is replaced before it leaves the server and restored on the way back. Whatever the recognition misses goes out unchanged.

- 

A log without content.

For every request to a language model in the chat and in the tools we record the time, the person, the provider, the model, the tokens and the cost. Every replacement is kept in a separate replacement protocol with the number of recognised values and checksums. We do not store the content.

- 

Vendor independent.

Today Cortecs in Vienna routes most requests to the models. A local model works just as well. The layer stays. The provider is replaceable.

 [See how this looks in production](https://ad-ze.de/en/erp-voice) 

Objections you are right to have

## Six objections against a small specialist provider. Our answer.

-  

Objection

„You're a small provider. What if you disappear tomorrow?"

Answer

For the results made for you, you get an exclusive, perpetual right of use. For reusable building blocks, you get a simple, perpetual right of use. A written handover process is contractually committed. AI-assisted development has clearly lowered the threshold for a takeover. A qualified vendor can take over an MCP-based layer, because the protocol is openly documented and the code is readable.

-  

Objection

„We need SLAs and guarantees, not trust."

Answer

Availability, response time, maintenance windows, escalation and backups are set out in the written contract. That includes clear terms and termination options.

-  

Objection

„Standard products are safer. Thousands of customers have already tested them."

Answer

That argument falls apart against the incidents documented above. EchoLeak, Slack AI, Replit — all standard products with millions of users. Thousands of testers did not prevent the gaps. For regulated industries, custom development with a clear boundary layer is no longer exotic in 2026; it's what auditors and insurers expect.

-  

Objection

„Why not wait until our ERP or CRM vendor solves this themselves?"

Answer

Because waiting costs money. The chart below shows it honestly. Your ERP vendor (abas, proALPHA, Sage, in-house) has AI modules on the roadmap. But concrete voice and audit layers will appear in one of the next major releases at the earliest, usually tied to an upgrade of your ERP contract. Your CRM vendor (HubSpot, Pipedrive, SevDesk) ships island solutions that don't talk to the ERP. Neither carries you across the actual workflow.

-  

Objection

„Where does this layer even sit? It sounds like yet another component."

Answer

Between two worlds that currently talk past each other: at the top, the wild AI skills of Glean, Copilot, ChatGPT — good for open search, bad for regulated workflows. At the bottom, hard-coded ERP APIs — exact but voiceless. The layer turns language, workflow and audit into one thing. It's a translation, not another island.

-  

Objection

„Who's liable if the layer itself fails?"

Answer

The contract sets this out. On top come professional liability insurance, automated tests in the code and version tags. In an incident we can see when which model was called for whom and how many values were replaced. We do not store the content.

The cost of waiting

## Opportunity costs, made concrete.

Three realistic company sizes. 1.8 hours per employee per workday lost to cross-system search. Conservatively, half of that is addressable. And: waiting doesn't cost linearly. Data silos solidify, workarounds proliferate, shadow AI takes hold, competitors capture share. 2% monthly compounding — the lines curve upward.

_Chart: cumulative delay cost over 36 months of waiting for three company sizes — 30, 100 and 300 employees — with 2% monthly compounding. At 30 employees this reaches roughly €1.85m after 36 months, at 100 employees roughly €6.18m, at 300 employees roughly €18.53m._

- 

Smaller operation — 30 employees

35.640 € / Monat · **1.853.079 €** nach 36 Monaten

- 

Mid-market — 100 employees

118.800 € / Monat · **6.176.931 €** nach 36 Monaten

- 

Larger operation — 300 employees

356.400 € / Monat · **18.530.792 €** nach 36 Monaten

Assumptions: 1.8 h/employee/workday lost (studies, see the ERP-voice-ready page), 50% of that addressable, 22 workdays/month, €60/h fully loaded, 2% monthly growth factor for compounding effects (data silos, shadow AI, market-share loss). No discounting. Follow-on costs from bad decisions on unfindable data come on top.

Honest framing

## What this page does not claim.

- The Garante fine against OpenAI was overturned in first instance. 

- Director external liability is not yet settled by the Federal Court. 

- Not every AI use case is high-risk. 

- A control layer is not risk-free. 

  ![Michael Schiller](https://ad-ze.de/rails/active_storage/representations/redirect/eyJfcmFpbHMiOnsiZGF0YSI6NDcwMzksInB1ciI6ImJsb2JfaWQifX0=--2f16e9a9191c77e10117f2790eed35b6118222e1/eyJfcmFpbHMiOnsiZGF0YSI6eyJmb3JtYXQiOiJ3ZWJwIiwicmVzaXplX3RvX2ZpbGwiOlsyMDAsMjAwXSwic2F2ZXIiOnsic3RyaXAiOnRydWUsInF1YWxpdHkiOjUwLCJpbnRlcmxhY2UiOnRydWUsIm9wdGltaXplX2NvZGluZyI6dHJ1ZSwiYWNjZXNzIjoic2VxdWVudGlhbCIsImJhY2tncm91bmQiOlsyNTUsMjU1LDI1NV19fSwicHVyIjoidmFyaWF0aW9uIn19--ddadceab840027df8f77d653b1b35d2b9a4a053b/image.webp?locale=en) 

## Let's talk about your case.

30 minutes. Your ERP, your risk profile, your question. No pitch.

 [Book a call](https://ad-ze.de/en/get-in-touch?service=operative-ai-safe) 

Sources and further reading

## Everything we claim has an address.

### Data protection and supervisory authorities

- [Garante per la Protezione dei Dati Personali — OpenAI decision, December 2024](https://www.garanteprivacy.it/home/docweb/-/docweb-display/docweb/10085432) — Press release, €15m fine — overturned by Tribunale Roma on 20 March 2026 
- [ANSA — Tribunale Roma overturns the OpenAI sanction (20 March 2026)](https://www.ansa.it/canale_tecnologia/notizie/tecnologia/2026/03/20/tribunale-roma-annulla-multa-da-15-milioni-di-euro-del-garante-privacy-a-openai_6560fadd-4099-451e-99a0-7fcbc9a285c1.html) — Ruling on the merits 
- [EDPB — Italian SA fines Luka Inc. (Replika), €5m](https://www.edpb.europa.eu/news/national-news/2025/ai-italian-supervisory-authority-fines-company-behind-chatbot-replika_en) 
- [Bloomberg — Samsung bans ChatGPT after internal leaks](https://www.bloomberg.com/news/articles/2023-05-02/samsung-bans-chatgpt-and-other-generative-ai-use-by-staff-after-leak) 
- [CNIL — Annual report 2024 (87 sanctions, over €55m)](https://www.cnil.fr/fr/rapport-annuel-2024) 

### EU AI Act

- [EU AI Act, Article 99 — Penalties](https://artificialintelligenceact.eu/article/99/) 
- [EU AI Act, Annex III — High-risk use cases](https://artificialintelligenceact.eu/annex/3/) 
- [EU AI Act, Article 12 — Record-keeping](https://artificialintelligenceact.eu/article/12/) 
- [EU AI Act, Article 13 — Transparency](https://artificialintelligenceact.eu/article/13/) 
- [EU AI Act, Article 14 — Human oversight](https://artificialintelligenceact.eu/article/14/) 
- [EU AI Act — Implementation timeline](https://ai-act-service-desk.ec.europa.eu/en/ai-act/timeline/timeline-implementation-eu-ai-act) 

### GmbH liability

- [Latham — OLG Dresden on personal director liability](https://www.lathamgermany.de/2022/02/olg-dresden-geschaftsfuhrer-haften-personlich-fur-datenschutzverstose-der-gmbh/) — Case ref. 4 U 1158/21 
- [Full text OLG Dresden, ruling of 30 Nov 2021](https://rewis.io/urteile/urteil/wrm-30-11-2021-4-u-115821/) 
- [§ 43 GmbHG — Director liability (German company law)](https://www.gesetze-im-internet.de/gmbhg/__43.html) 

### Documented AI security incidents

- [HackTheBox — EchoLeak / CVE-2025-32711 (M365 Copilot, zero-click)](https://www.hackthebox.com/blog/cve-2025-32711-echoleak-copilot-vulnerability) 
- [PromptArmor — Slack AI data exfiltration via indirect prompt injection](https://www.promptarmor.com/resources/data-exfiltration-from-slack-ai-via-indirect-prompt-injection) 
- [The Register — Slack AI prompt injection](https://www.theregister.com/2024/08/21/slack_ai_prompt_injection/) 
- [Fortune — Replit AI wipes production database](https://fortune.com/2025/07/23/ai-coding-tool-replit-wiped-database-called-it-a-catastrophic-failure/) 
- [AI Incident Database — entry #1152 (Replit)](https://incidentdatabase.ai/cite/1152/) 
- [Heise — Microsoft Copilot falsified access logs](https://www.heise.de/en/news/AI-assistant-Microsoft-s-Copilot-falsified-access-logs-for-months-10578970.html) 
- [Simon Willison — Lethal Trifecta](https://simonwillison.net/2025/Jun/16/the-lethal-trifecta/) — Conceptual reasoning 

### Standards and architecture

- [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/resource/owasp-top-10-for-llm-applications-2025/) 
- [NIST AI 600-1 — Generative AI Profile](https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-generative-artificial-intelligence) 
- [ISO/IEC 42001:2023 — AI Management System](https://www.iso.org/standard/42001) 
- [ISO/IEC 27001 — Information Security Management](https://www.iso.org/standard/27001) 
- [Model Context Protocol — Authorization spec](https://modelcontextprotocol.io/specification/2025-11-25) 

### Market reality

- [Gartner — Over 40% of agentic-AI projects will be cancelled by end of 2027](https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027) 
- [Computerworld — Microsoft addresses Copilot oversharing (40% rollouts delayed)](https://www.computerworld.com/article/3616459/microsoft-moves-to-stop-m365-copilot-from-oversharing-data.html) 
- [CIO.com — SAP's new Joule pricing neither transparent nor explainable](https://www.cio.com/article/655478/saps-new-generative-ai-pricing-neither-transparent-nor-explainable-yet.html) 
- [Microsoft — EU Data Boundary completed, February 2025](https://blogs.microsoft.com/on-the-issues/2025/02/26/microsoft-completes-landmark-eu-data-boundary-offering-enhanced-data-residency-and-transparency/) 
- [Cybernews — Flex Routing outside the EU remains possible](https://cybernews.com/ai-news/microsft-copilot-eu/) 

As of 21 May 2026. All sources verified at the time of publication. URLs may change.

  ![Michael Schiller](https://ad-ze.de/rails/active_storage/representations/redirect/eyJfcmFpbHMiOnsiZGF0YSI6NDcwMzksInB1ciI6ImJsb2JfaWQifX0=--2f16e9a9191c77e10117f2790eed35b6118222e1/eyJfcmFpbHMiOnsiZGF0YSI6eyJmb3JtYXQiOiJ3ZWJwIiwicmVzaXplX3RvX2ZpbGwiOlsyMDAsMjAwXSwic2F2ZXIiOnsic3RyaXAiOnRydWUsInF1YWxpdHkiOjUwLCJpbnRlcmxhY2UiOnRydWUsIm9wdGltaXplX2NvZGluZyI6dHJ1ZSwiYWNjZXNzIjoic2VxdWVudGlhbCIsImJhY2tncm91bmQiOlsyNTUsMjU1LDI1NV19fSwicHVyIjoidmFyaWF0aW9uIn19--ddadceab840027df8f77d653b1b35d2b9a4a053b/image.webp?locale=en) 

## Let's talk about your case.

30 minutes. Your ERP, your risk profile, your question. No pitch.

 [Book a call](https://ad-ze.de/en/get-in-touch?service=operative-ai-safe) 

30 minutes on your risk

 [Book a call](https://ad-ze.de/en/get-in-touch?service=operative-ai-safe)
