Why AI Is a Sustainability Management Requirement — Clause 10.11

Quick Read

AI governance is now a core requirement of SPK CSMS1000:2026 because AI systems are embedded throughout sustainability management—from GHG calculation tools to supplier screening platforms—and their quality, bias, and reliability directly affect the credibility of sustainability data and reporting. The standard treats AI tools as part of the data supply chain subject to internal control and sustainability risk assessment, closing a gap where organizations might have strong human controls but no governance over the algorithms processing their sustainability information. Beyond data integrity, AI systems themselves carry sustainability and ethical implications—energy consumption, algorithmic bias, privacy risks—that organizations must address within their management system.

Executive Summary

The inclusion of AI governance in SPK CSMS1000:2026 surprises many practitioners. Sustainability management and artificial intelligence seem like separate disciplines. They are not. AI systems are already deeply embedded in sustainability management — calculating GHG inventories, screening suppliers for ESG risk, generating sustainability reports, processing environmental sensor data, and scoring ESG performance. The quality, reliability, and ethical dimensions of these AI systems directly affect the credibility of the CSMS they support.

This paper explains why AI governance is a sustainability management system requirement, what Clause 10.11 requires, how it aligns with ISO 42001, and what organisations need to build to demonstrate AI governance in a Speeki Meridian™ engagement.

If an AI system calculates your Scope 3 emissions, screens your suppliers for human rights risk, or generates the data that appears in your sustainability report — the quality and governance of that AI system is a sustainability management system concern. Clause 10.11 addresses this directly.

1. Why AI Governance Belongs in a Sustainability Management System

1.1 AI in sustainability management

Most large organisations now use AI in sustainability contexts, whether or not they recognise it as such. Common applications include: automated GHG inventory calculation tools using AI to estimate activity data and apply emission factors; supplier ESG screening platforms using machine learning to assess supplier risk from public data; sustainability reporting platforms using AI to generate narrative from data; environmental monitoring systems using AI to process sensor data; and carbon accounting software using AI to categorise and classify expenditure for Scope 3 calculation.

Each of these applications creates AI governance requirements. If the GHG calculation tool applies incorrect emission factors or uses a flawed algorithm, the Scope 1 and 2 figures in the sustainability report are wrong. If the supplier screening platform has biases in its training data, the supply chain risk assessment may systematically underestimate risk in certain geographies. If the carbon accounting software miscategorises expenditure, the Scope 3 Category 1 figure is unreliable.

1.2 AI as an ICSR risk

From an ICSR perspective (Clause 10.4), AI systems used in sustainability data processing are part of the data supply chain. ICSR controls must address the AI tools that process or calculate sustainability data — not just the human processes that collect and review it. An organisation that has robust ICSR controls over its human data collection processes but no governance over the AI systems that process that data has an ICSR gap.

1.3 AI and ethical dimensions of the CSMS

Beyond data quality, AI systems deployed by the organisation may have their own sustainability impacts and ethical implications: energy consumption of large AI models, bias in AI-assisted HR decisions, privacy implications of AI surveillance, and the environmental footprint of AI infrastructure. These are sustainability topics that the CSMS may need to manage — particularly for organisations in the technology sector where AI products are central to the business model.

2. What SPK CSMS1000:2026 Requires

Clause 10.11 requires AI governance aligned with ISO 42001:2023 — the international standard for AI management systems. The clause applies to material AI systems: those that have significant influence on the organisation's sustainability performance, sustainability data quality, or sustainability-related decisions.

2.1 AI system inventory

The organisation must maintain a documented inventory of AI systems that are material to the CSMS. The inventory must identify: the system and its purpose; the sustainability function it supports; the data it processes or outputs it produces; the organisation's role (developer, deployer, or user); and the risk level of the system based on its influence on material sustainability data or decisions.

2.2 Risk assessment for material AI systems

For each material AI system, the organisation must conduct and document a risk assessment covering: data quality risks (bias, incompleteness, or inaccuracy in training data or input data); model performance risks (accuracy, reliability, and appropriate use within the system's design parameters); transparency risks (explainability of AI outputs, particularly for high-stakes sustainability decisions); and ethical risks (fairness, privacy, and human rights implications of AI deployment).

2.3 Controls for material AI systems

Controls must address the identified risks. For AI systems used in sustainability data processing, controls connect directly to ICSR (Clause 10.4): the AI system's calculation methodology must be documented; outputs must be validated against independent data sources; and changes to the AI system (model updates, algorithm changes, data source changes) must be managed through change control.

2.4 Human oversight

The standard requires that material AI systems are subject to appropriate human oversight — particularly for AI outputs that inform material sustainability decisions or that appear in sustainability disclosures. AI-generated sustainability data must be reviewed by humans with the competence to assess its reasonableness before it is used in reporting.

Speeki Meridian™ — Auditor Expectations

At Stage 1, assessors will request the AI system inventory and evidence that material AI systems have been identified. For organisations that use AI in GHG calculation, supplier screening, or sustainability reporting, the absence of an AI inventory is a Stage 1 finding. At Stage 2, assessors will select the AI systems most material to sustainability data quality and test: Is there documentation of the system's methodology? Has the output been validated against independent data? Is there a change control process for system updates? Is there human review of AI outputs before use in reporting? For organisations that develop AI products or services (technology sector), assessors will also test whether the sustainability impacts and ethical dimensions of those AI products are addressed in the CSMS — not just the AI systems used internally for sustainability management.

Implementation Guidance

Start with the inventory. Identify every software system, algorithm, or automated tool that contributes to your sustainability data or informs sustainability decisions. Include: your carbon accounting software, your ESG reporting platform, your supplier risk screening tool, your energy management system, any AI-generated data in your sustainability report. This inventory is almost certainly longer than you expect. For each system, assess materiality: does this system's output materially affect reported sustainability data or material sustainability decisions? If yes, it is material and requires risk assessment and controls. For AI systems used in GHG calculation, connect the governance documentation directly to your ICSR controls. The calculation methodology of the AI system is part of the ICSR documentation — assessors will ask for it during a data walk-through.

About Speeki

Speeki is an accredited certification body operating across more than 100 countries. Speeki certifies organisations against SPK CSMS1000:2026 through the Speeki Meridian™ certification programme. Speeki is a certification body — it does not provide sustainability consulting or advisory services of any kind.

For current details of Speeki's accreditations, scope of certification, and service offerings, visit speeki.com. You can also ask Nicole AI on the Speeki website to find the information you need.