Services/Service 05Tool
AI-Assisted Green Claims Screening
Built by Regulatory Specialists. Trained on the Right Risks.
Most AI tools in the compliance space are built for general legal risk. They flag broad categories of concern, produce outputs that require interpretation, and may not be calibrated to the specific standards applied to environmental claims.
This tool is different because it was built on more precise foundations.
Verdant ESG's AI screening tool is developed and updated by specialists who work daily with the CMA Green Claims Code and ASA/CAP environmental advertising guidance. It scans for the specific language patterns, structural weaknesses, and evidentiary gaps that regulators and complaint handlers search for.
It is not meant to replace legal advice, a structured audit, or human judgement. It is the first pass that sets the tone for speed, consistency, and calibration, ensuring that, as the volume of work increases, nothing is missed.
The problem
The Problem It Solves
In most organisations, environmental claims are created quickly. Marketing drafts copy, briefs to an agency, product description updates, social media posts, and reports might all be done under pressure.
At each of these points, the question of whether the language used meets the CMA and ASA/CAP standards is either asked informally, not asked at all, or deferred to a review process that may not have the bandwidth to catch issues.
The result is not malicious or structural, but claims accumulate across channels faster than compliance functions can review them. This creates new ground for risk.
The AI screening tool addresses that structural gap, processes the language quickly, flags specific indicators that may invite scrutiny, and produces a report with recommendations for what to review and why.
Capabilities
What the Tool Scans For
The tool is designed to identify common risk categories across ASA adjudications, CMA enforcement guidance, and the Green Claims Code.
- 01
Vague Environmental Terminology
Language that implies broad environmental benefits without defined scopes, basis, or limitations is flagged. This could be terms used without qualification (sustainable, eco-friendly, green, natural, etc.) or necessary evidence. They aren't flagged automatically as non-compliant but as requiring substantiation before publication.
- 02
Absolute Claims
Statements presented in absolute terms are likely to need absolute proof. The tool identifies the framing and flags what calls for high evidentiary standards to support.
- 03
Carbon and Net Zero Language
Carbon neutrality and net-zero claims receive some of the highest scrutiny under current enforcement. The tool identifies and flags specific requirements such as boundary definition, methodology transparency, third-party verification, and honest reporting of emissions. Claims lacking these elements present high regulatory risk.
- 04
Unsupported Comparative Claims
Language that implies a product or service is better, greener, cleaner, or more sustainable than it is, with no basis, violates the ASA/CAP framework. The tool identifies comparative framing and flags where the basis for comparison is unclear or missing.
- 05
Supply Chain Assertions
The tool finds claims that depend on supplier-level evidence (such as sourcing statements, manufacturing claims, ethical language, etc.) where the evidentiary chain of custody may be incomplete. The tool flags statements that need supply chain evidence review.
- 06
Missing or Inadequate Qualifiers
Which of your claims would be defensible with appropriate qualification, but currently don't have it? The tool identifies scope limitations, geographic qualifiers, time boundaries, methodology references, etc., to convert a high-risk claim into a defensible one.
- 07
Life-cycle Red Flags
Single-attribute claims that obscure negative impacts elsewhere in a product's life cycle are risky. The CMA's fifth principle requires claims that account for the full picture. The tool flags where framing may create misleading impressions overall.
- 08
Regulatory Watchlist Terminology
The tool maintains a list of specific terms and phrases identified as problematic in ASA adjudication, CMA guidance, or enforcement action. This list is updated as regulatory changes happen.
The output
What Does The Report Look Like?
The tool's review returns a risk report organised by claim. For each flagged item, you get:
The report is designed to help with decision-making and action by marketing, ESG, or legal teams without a specialist on hand for every flag. It tells you what the problem is, why it falls in that category, and what actions to take.
Sample report structure
- 1
The specific language identified
- 2
The risk category it falls into
- 3
The regulatory basis for its classification as a risk (based on the CMA and ASA/CAP code)
- 4
The evidence needed to support the claim
- 5
Recommended actions (such as substantiation review, qualification, rewrite, or referral to legal or compliance)
The limits
What This Tool Does Not Do
It Does Not Assess Your Evidence
The tool reviews language. It cannot access your substantiation files, supplier documentation, or certification records. A claim that passes the screening tool is not confirmed as compliant, just that it doesn't have obvious language-level indicators. Evidence review is a whole other step.
It Is Not a Replacement For Legal Advice
Where a claim turns out to have legal risk, particularly if it is part of an existing complaint, enquiry, or high-value commercial commitment, legal advice is indispensable. The screening tool only offers pre-publication review, not legal opinions.
It Does Not Replace Human Judgement
The tool is calibrated to the regulatory frameworks we work with on a daily basis, but context is crucial. A claim that is flagged as high risk in isolation might be defensible in context, with the right evidence. The tool's report is a starting point for human review, not a compliance verdict.
It Does Not Create an Audit Trail by Itself
The tool's output should be documented as part of a structured internal claims approval process, alongside the evidence packs and required sign-offs. The tool supports this process rather than constituting it.
Workflow
How Does The Tool Fit Into Your Claims Process
The screening tool is designed to be at the front end of your internal claims approval workflow. It comes after drafting but before substantiation review and sign-off.
- Step 01
Draft
- Step 02
AI Screening
- Step 03
Substantiation Review
- Step 04
Legal / Compliance Sign-Off
- Step 05
Publication
- Step 06
Archiving for Evidentiary and Auditing Purposes
At the draft stage, AI screening can catch high-risk language before it reaches the review stage, reducing the workload on legal and ESG teams. It also ensures that the claims entering the review phase are cleaner.
If you deploy it as a standing auditing tool, it can be run periodically across existing published content to identify claims that may expose you to legal issues. It could be that evidence has changed, certifications have lapsed, or regulatory guidance has evolved.
Whatever the reason for your exposure, you don't want to miss it.
Who this is for
Who This Is For
- 01
Marketing and communications teams who need a fast, structured check before forwarding material to agencies or approving copy.
- 02
In-house legal and compliance functions that deal with high volumes of claims across multiple teams and channels.
- 03
ESG leads responsible for sustainability reports, investor communications, and public-facing commitments.
- 04
Agencies developing sustainability campaigns that want to find and mitigate client-side risk in their briefs.
- 05
Businesses in high-scrutiny industries, such as fashion, FMCG, financial services, food, and energy. The volume of publications tends to be high and regulatory attention is always on.
The Principle Behind The Tool
Regulators almost never need to review every claim a business has ever made. They usually focus on the most current vague, unsubstantiated, or misleading claims. In a landscape where most are publishing environmental claims across multiple channels and formats, the likelihood that one of them could be in violation is not far-fetched.
The screening tool is built on the idea that the most protected businesses are those that looked at their own language the way a regulator would — before a regulator did.