WasteTrade Group

AI Use Policy

Version 1.1 — Effective July 2026 · Applies to the WasteTrade Platform and to all Personnel of the WasteTrade Group

This policy explains how artificial intelligence is used across the WasteTrade Platform and by the personnel of the WasteTrade Group. It is referenced in Section 12 of the Group Code of Conduct and in the Privacy Notice and Data Use Clause, and should be read together with the Platform Terms and Conditions v2.7 (the “Terms”), in particular clauses 5.2, 7.16, 16.2 and 16.3.

Issuing entity: Waste Trade Holdings Limited (trading as WasteTrade), Company No. 13314586, England & Wales, on behalf of the WasteTrade Group (Waste Trade Holdings Limited; WasteTrade.com Romania S.R.L., Trade Register J2025037071001, CUI 51846722; and WasteTrade Ghana Holdings Limited, Company Registration No. ET000430425).

Part A — How the Platform uses AI

The Platform uses automated and semi-automated systems, including artificial intelligence and machine learning, in the following functions:

Function	Description
Classification and categorisation	Assisting the classification and categorisation of listings and materials from User inputs and public data. Users remain solely responsible for the accuracy and regulatory validity of all classifications and descriptions (clause 5.2 of the Terms).
Search ranking and matching	Ranking search results and suggesting potential counterparties or materials. These are commercial introduction tools only (clause 7.16 of the Terms).
Lead scoring and routing	Prioritising and routing enquiries and leads to the commercial team.
Price indication	Producing indicative pricing analytics from aggregated market data. Indicative only; contract prices are those in the Transaction documentation.
Fraud, sanctions and compliance screening support	Supporting KYC, sanctions and fraud screening. Screening outcomes affecting a User are reviewed by a human before action is taken.
Document extraction	Extracting data from uploaded documents (weighbridge tickets, certificates, transport documents) to populate records.
Assistant and chatbot features	Answering queries and assisting navigation. Users are told when they are interacting with an AI system.

What Platform AI is not. AI outputs do not constitute certification, endorsement, verification or warranty by WasteTrade; they do not verify regulatory status; and they do not represent a match against any Buyer’s authorisation scope, including any Food-Contact Recycling Authorisation (clauses 5.2 and 7.16 of the Terms). WasteTrade personnel do not present AI outputs to counterparties as certification.

Human in the loop. People, not systems, are accountable for decisions affecting Users. Decisions that materially affect a User — refusal of registration, account suspension or termination, cancellation of a Transaction, or a compliance escalation — involve human review and are not taken solely by automated means.

Part B — Internal use of AI by Personnel

Personnel may use AI tools approved by the Group, including Anthropic Claude (including the Group’s internal Claude-based assistants), Salesforce Einstein, and such other tools as the AI Lead approves from time to time. The following rules apply to all internal AI use:

1. Confidential information, counterparty data and personal data must not be entered into unapproved AI systems.

2. AI outputs are drafts and analysis, not decisions. Every output used in commercial documentation, classification, compliance or counterparty communication is reviewed by a responsible person before use, and that person is accountable for it.

3. AI must not be used to generate or manipulate evidence, certificates, photographs, weighbridge data or other records. Falsification by any means is a fundamental breach of the Terms (clause 5.3) and gross misconduct under the Code of Conduct.

4. Commercial documents produced with AI assistance follow the same document discipline as all others: template control, dual review, director or Sales Director approval, PDF-only issue, and same-day filing.

5. Personal data used with AI tools is handled in accordance with the Privacy Notice and Data Use Clause and applicable data-protection law.

Part C — Responsible AI principles

WasteTrade’s use of AI is guided by the following principles:

Human accountability. People, not systems, are accountable for decisions affecting Users. AI informs decisions; it does not replace human judgement on matters that materially affect a User.

Transparency. Users are told when they are interacting with an AI system, and the existence of automated processing is disclosed in the Terms and Conditions and the Privacy Notice.

Fairness. WasteTrade takes reasonable steps to avoid AI outputs that unfairly disadvantage Users, and reviews AI-assisted processes where bias is a risk.

Data protection. Personal data used in AI systems is handled in accordance with the Privacy Notice and applicable data-protection law, including the right of an identifiable natural person to opt out of the use of their personal data for model training under clause 16.3 of the Terms (personal data only; it does not require the recall, deletion or re-training of models trained before the request).

Security. AI tools are used in a manner consistent with WasteTrade’s confidentiality and information-security obligations.

Proportionality and lawfulness. WasteTrade uses AI where it improves the service and operates within applicable law, including the EU AI Act (Regulation (EU) 2024/1689), data-protection law, and the Platform-to-Business Regulation (Regulation (EU) 2019/1150).

Part D — Governance

The Group’s AI Lead is the Chief Operating Officer. The AI Lead maintains the list of approved tools, reviews this policy at least annually and on any material change to the systems in use or to applicable law, and is the escalation point for AI-related concerns. Questions about this policy should be directed to the AI Lead via privacy@wastetrade.com.

Related documents: Platform Terms and Conditions v2.7; Privacy Notice and Data Use Clause; Group Code of Conduct.

WasteTrade AI Use Policy — Version 1.1 — Effective July 2026