The Accounting & Audit Industry Framework

Accounting and audit firms operate at the intersection of three pressures: client demand for AI-enabled efficiency in the conduct of substantive procedures, regulatory exposure under the EU AI Act and the International Standards on Auditing, and the institutional duty of professional scepticism. The Raydorf Accounting and Audit Framework adapts the seven dimensions of the Standard to these pressures.


§ I Evaluation

What the Accounting & Audit Framework evaluates

  1. Strategy & Leadership

    Engagement leadership accountability. AI strategy reviewed by the firm leadership and quality management function on a defined cadence. A named individual responsible for the firm's AI programme, integrated with the firm's system of quality management under ISQM 1.

  2. Governance, Risk & Compliance

    Written AI policy aligned with the International Standards on Auditing, applicable ethical standards (IESBA Code), and the EU AI Act. Risk classification of AI use cases against the auditor's independence obligations. Engagement letter language addressing AI involvement. Quality review procedures extended to AI-assisted work.

  3. Data & Knowledge Infrastructure

    Engagement-level sensitivity labelling. Retention discipline aligned with the audit working-paper requirements at ISA 230. A queryable institutional knowledge base populated through a defined engagement-close process. Separation of client data across engagements maintained in the knowledge architecture.

  4. Workflow Redesign & Operations

    Identified workflows where AI is the default first pass — typically journal entry testing, transaction analysis, contract review, risk assessment under ISA 315 (Revised), and analytical procedures under ISA 520. Documented before-and-after metrics tied to audit quality indicators.

  5. Talent & Operating Model

    AI fluency across audit and assurance professionals and operational staff. A designated operator role for the firm's AI infrastructure. Hiring and progression criteria that reflect AI-era competencies. Defined training cadence covering both capability and the specific obligations of professional scepticism.

  6. Client Experience

    Disclosure practice for AI involvement in audit and assurance engagements. Client-facing AI features where appropriate. Demonstrable service improvements attributable to AI maturity, reported in a manner consistent with the firm's independence obligations.

  7. Measurement & Accountability

    Audit quality metrics tied to AI use. Logged human oversight on AI-assisted procedures, in a form that supports the engagement partner's sign-off under ISA 220 (Revised). An audit trail sufficient to reconstruct AI involvement on any engagement, at any time, in support of inspection by audit oversight bodies.

§ II EU AI Act

EU AI Act considerations specific to audit practice

Most uses of AI in audit and assurance practice fall outside the EU AI Act's high-risk categories. The framework engages the high-risk regime more frequently than the legal sector, however, because audit firms may serve clients whose own AI systems are high-risk under Annex III, and because certain assurance engagements that touch the assessment of natural persons (for example, anti-money-laundering reviews, fraud-risk scoring, or assurance over creditworthiness models) may themselves engage the regime. The framework includes screening criteria to surface these uses, and the EU AI Act Readiness Attestation evaluates the firm's response to them.