ISEP's New Guidance on Responsible AI Use in Sustainability Assessments
Artificial intelligence tools are now common across environmental and sustainability work. Professionals use them to analyse data, summarise research, and speed up report writing. However, the Institute of Sustainability and Environmental Professionals has now published formal guidance on how these tools should be used in environmental impact assessment. The message is clear: AI can help, but it must not replace professional judgement or weaken accountability.
The guidance sets out six principles. It calls for transparency, expert review, and legal compliance. Importantly, it also reminds practitioners that they remain fully responsible for any work produced with AI assistance. This matters because environmental impact assessments carry legal weight and inform planning decisions that affect communities, habitats, and infrastructure projects.
For sustainability teams and consultancies working on impact assessments, the guidance offers a practical framework. It clarifies where AI fits and where it does not. Consequently, businesses now have a clearer basis for updating internal policies and training staff on responsible use.
What the ISEP guidance covers
The advice note, titled "Using AI in EIA," treats artificial intelligence as a support tool rather than a decision-making system. It emphasises that users must understand how these tools work, what risks they carry, and what obligations apply when handling sensitive or proprietary information. Moreover, it stresses that outputs must align with established assessment frameworks, both national and international.
Transparency is a central requirement. The guidance calls for full disclosure in reports, including the name of the AI tool used, the date of use, and a description of how and where it contributed to the document. This level of detail is intended to maintain trust and enable scrutiny by regulators, consultees, and decision-makers.
Verification is equally important. Any content generated or supported by AI must be checked against evidence and reviewed by qualified professionals. This step is non-negotiable, according to the guidance, because AI systems can produce plausible-sounding material that contains errors or unsupported claims.
Six principles for responsible AI use
The guidance is built around six core principles. First, professionals must retain full accountability for AI-assisted work. The tool may assist, but responsibility for accuracy and compliance stays with the practitioner. Second, outputs must fit within regulatory frameworks and recognised assessment standards. AI cannot be allowed to bypass established methodologies.
Third, transparency requires open statements about what type of AI was used and to what extent. Fourth, verification and peer review are mandatory. AI outputs must be cross-checked against evidence and examined by experts. Fifth, data quality matters. Poor inputs lead to poor outputs, so the quality of prompts and source data is critical.
Sixth, AI should complement professional judgement, not substitute for it. The guidance describes AI as a utility tool. It can save time and handle repetitive tasks, but it cannot replace the expertise, context, and critical thinking that experienced practitioners bring to environmental assessment work.
ISEP also highlights practical barriers to wider adoption. These include the need for governance structures, audit trails, and consistent human oversight. Without these safeguards, the risks of misuse or error increase significantly.
Why environmental professionals need clearer AI rules
Environmental impact assessments are not routine documents. They inform planning decisions, support regulatory approvals, and can be challenged in court. Therefore, the evidence base must be robust, traceable, and legally defensible. AI tools can speed up certain tasks, but they also introduce new risks.
One risk is over-reliance. If practitioners treat AI as a shortcut, they may overlook errors, fail to interrogate assumptions, or miss context that a machine cannot recognise. Another risk is confidentiality. Many AI tools process data externally, which raises concerns about how sensitive project information is stored or used.
There is also the problem of hallucinations. AI systems sometimes generate content that appears credible but contains factual errors or invented references. In environmental work, where evidence standards are high, this is a serious concern. Consequently, expert review is essential at every stage.
The ISEP guidance addresses these risks by setting clear expectations. It does not ban AI use. Instead, it establishes boundaries and responsibilities. Professionals can use the tools, but they must do so competently, transparently, and with full accountability.
Disclosure, audit trails, and legal compliance
The guidance states that AI tools "should only be used when their application complies with legal requirements and when users have a clear understanding of how to use them effectively, efficiently and ethically." This wording is significant. It places the burden on the user, not the tool, to ensure compliance.
Transparency is not optional. Reports must include full and open statements about the type of AI used and the extent of its involvement in assessment and reporting. This requirement serves two purposes. It allows reviewers to understand how conclusions were reached, and it provides an audit trail if questions arise later.
For businesses, this means documentation becomes more detailed. Teams will need to record which tools were used, what prompts were entered, and how outputs were verified. This adds a layer of administrative work, but it also strengthens the defensibility of the final report.
Legal compliance is another key concern. Environmental regulations vary by jurisdiction, and assessment standards differ between sectors. AI tools do not inherently understand these nuances. Therefore, professional oversight is required to ensure that outputs meet the relevant legal and technical standards.
Core requirements for sustainability teams
Here are the essential points for any organisation working on environmental impact assessments:
- Professionals remain fully accountable for all AI-assisted work, regardless of the tool used.
- AI outputs must align with established regulatory frameworks and recognised assessment methodologies.
- Reports must disclose where and how AI was used, including tool names, dates, and the nature of its contribution.
- All AI-generated content must be verified against evidence and reviewed by qualified experts before publication.
- Data quality is critical because poor inputs produce unreliable outputs that undermine the credibility of the assessment.
- AI should support professional judgement, not replace it, and must not become a substitute for expertise and critical thinking.
- Confidentiality and intellectual property obligations apply when using third-party AI tools, particularly those that process data externally.
Practical steps for businesses using AI in sustainability work
Companies and consultancies will need to review their internal policies in light of the ISEP guidance. Many organisations already use AI tools for research, drafting, and data analysis. However, few have formal governance in place to manage the risks that come with this use.
First, businesses should define approved use cases and prohibited applications. For example, AI might be acceptable for summarising research but not for generating technical conclusions without expert review. Clear boundaries help prevent misuse and ensure consistency across teams.
Second, mandatory human review should be built into workflows. This means assigning qualified professionals to check all AI-assisted content before it is finalised or submitted. Review should focus on accuracy, compliance, and alignment with project-specific requirements.
Third, reporting and audit requirements need to be formalised. Teams should document what tools were used, what inputs were provided, and how outputs were verified. This creates a clear trail that supports transparency and accountability.
Fourth, data protection and confidentiality safeguards must be strengthened. Many AI tools operate in the cloud and process data externally. Businesses need to understand where data goes, how it is stored, and whether it is used to train the AI system. For sensitive projects, this may rule out certain tools entirely.
Finally, training is essential. Staff need to understand the limitations of AI, the risks of over-reliance, and the professional standards that apply to their work. Training should cover practical topics such as prompt design, output verification, and regulatory compliance. It should also address ethical considerations, including bias, fairness, and the importance of maintaining public trust.
For companies looking to build capability in this area, SBS Academy training on environmental reporting and compliance provides practical guidance for sustainability teams navigating these new tools responsibly.
Balancing efficiency with responsibility
The professional conversation around AI in environmental work is shifting. Early enthusiasm focused on speed and efficiency gains. Now, the focus is on governance, quality control, and accountability. This reflects a maturing understanding of what AI can and cannot do.
AI tools excel at certain tasks. They can process large datasets, identify patterns, and generate draft text quickly. They can also help non-specialists access technical information more easily. However, they cannot replace the judgement, experience, and contextual understanding that human professionals bring to complex environmental questions.
For sustainability practitioners, this means AI is best used as an assistant, not an author. It can support research, suggest structures, and flag issues for further investigation. It cannot make decisions, assess trade-offs, or apply nuanced regulatory requirements without human direction.
The ISEP guidance reflects this balanced view. It encourages adoption where AI adds value, but insists on safeguards to protect quality, compliance, and trust. Consequently, businesses that adopt AI responsibly will gain efficiency without compromising the rigour that regulators and stakeholders expect.
Organisations seeking support with ESG compliance and environmental reporting can benefit from structured guidance on integrating new tools while maintaining robust governance and audit trails.
Where to find the ISEP guidance and related resources
The full ISEP advice note, "Using AI in EIA," is available from the Institute of Sustainability and Environmental Professionals. It provides detailed recommendations for practitioners working on environmental impact assessments and related sustainability reporting.
For broader context on environmental assessment frameworks, the UK government's planning guidance offers authoritative information on regulatory requirements and standards. Businesses involved in environmental work should also consult sector-specific guidance from regulators such as the Environment Agency.
Professional bodies including the Institute of Environmental Management and Assessment provide resources on emerging practice in environmental assessment. These organisations regularly update their guidance to reflect changes in technology, regulation, and professional standards.
Finally, businesses preparing environmental reports or sustainability disclosures may find it helpful to review the UK government's net zero strategy and related policy documents, which set out the wider regulatory context for environmental assessment and climate-related reporting.