Disclaimer

The content of this website is published for general information, portfolio presentation and service description. It is not legal, financial, medical or tax advice, and it should not be relied on as a substitute for advice from a qualified professional in the relevant field. Where a question touches Ghanaian law — including Data Protection Act, 2012 (Act 843) — consult a qualified Ghanaian lawyer.

Information is published in good faith and I take care to keep it accurate, but I make no guarantee that it is complete, current or free from error. Any decision you take based on this content is your own.

On service pages I describe capabilities and the problems they are built to solve. I do not publish guaranteed outcomes, and you should be sceptical of anyone who does — including me, if I ever did. In particular:

• No guaranteed conversion rate, revenue figure, cost saving or sales uplift. Commercial results depend on your product, pricing, market, traffic and sales process, none of which I control.
• No guaranteed response-time or accuracy figure for an AI system. Any figure shown is a design target agreed for a specific build, not a promise of what your system will do.
• No claim that AI output is error-free. Language models are probabilistic and will occasionally produce something wrong, confident-sounding and incorrect.
• No claim of certification, accreditation or regulatory approval. Where a status is pending — such as my own registration on the Data Protection Register — I say so plainly rather than implying it is held.
• No claim that data can never be compromised. What I commit to is the documented set of safeguards described in the Privacy Policy, plus a breach-notification procedure under section 31 of the Act.

Because a substantial part of what I now build involves artificial intelligence, the limitations deserve stating separately rather than hiding in general terms.

• Outputs are probabilistic. A chatbot can misunderstand a question, miss context, or invent a detail that sounds authoritative and is not in your source material. I design for escalation to a human and for honest uncertainty, but the underlying behaviour cannot be engineered away.
• Knowledge is bounded by what the system was given. If a bot does not know something, it may still answer plausibly. Grounding the system in your own documents, and testing it against real questions, is the mitigation — not a guarantee.
• Retrieval quality depends on your source material. Duplicated, contradictory, outdated or poorly structured documents produce correspondingly poor answers, no matter how good the search layer is.
• Third-party model and platform providers set the limits. Pricing, availability, rate limits, moderation behaviour and terms can change without notice, and are outside my control. I build so providers can be swapped, and I disclose where that exposure sits.
• Automation inherits upstream failures. If a system an automation reads from is down, delayed or returns malformed data, the automation will behave incorrectly. I build in validation, alerting and failure handling, but I cannot remove the dependency.
• Automation that changes data can be destructive at scale. I design reversibility, dry-run capability and audit logging, but the responsibility for what a live automation writes in production remains with the client as data controller.
• Human review is not optional in high-stakes domains. For medical, financial, legal, employment or safeguarding decisions, an automated system must not be the final decision-maker. Section 41 of the Act gives data subjects rights in relation to automated decision-making, which is a legal obligation rather than a matter of taste.

This website links to external sites that are not provided, maintained or affiliated with me. I have not reviewed them, I do not control their content, availability or privacy practices, and the inclusion of a link is not a recommendation or endorsement. Use them at your own risk.

Technologies, platforms and providers referenced on this site — including hosting, databases, payment gateways, messaging platforms, model providers and third-party APIs — belong to their respective owners. Naming a tool describes how a system may be built; it is not a claim of partnership, sponsorship or endorsement.

Projects shown in the portfolio section are presented for demonstration. Where a project is described:

• Features and behaviour may have changed since publication.
• A project may be ongoing, in maintenance, or discontinued.
• Third-party services or APIs it relies on may have changed, been repriced or been withdrawn.
• Screenshots and descriptions are representative and may not reflect the current production state.
• Project availability, performance and functionality are not guaranteed.
• Client names, logos and identifying details appear only with the client's permission.

AI services are currently presented as capabilities rather than shipped case studies, because there are no completed client AI engagements to reference. When one ships it will be added here as a case study, on the same terms as any other.

I aim to keep this website available, but I do not guarantee it will be uninterrupted, timely, or free of errors. The site may be unavailable because of deployment work, maintenance, host or upstream provider problems, or internet connectivity issues outside my control. I do not warrant that the website will be error-free or continuously available, and I do not offer a service level agreement on the website itself.

The Privacy Policy describes my actual practices and actual technical measures, and maps them to the eight principles in section 17 of the Data Protection Act, 2012 (Act 843). It is not a compliance certificate. Compliance with the Act is assessed by the Data Protection Commission of Ghana, and my own registration on the Data Protection Register is currently in progress. I would rather tell you that accurately than present a claim I cannot evidence.

You are responsible for:

• Deciding whether to act on the information here, and for the consequences of that decision.
• Providing accurate, complete and lawful information, and for holding the rights to what you provide.
• Reviewing anything an AI or automation system produces before it affects a customer, an employee or a third party.
• Complying with Ghanaian law in your use of anything I build, including obligations you retain as data controller.
• Maintaining the accuracy of contact details you give me, so that I can reach you about a matter affecting your data.

To the fullest extent permitted by law, Daniel Amekpoagbe is not liable for any loss or damage, indirect or consequential, arising from:

• Loss of data, revenue or profit
• Business interruption
• Use of, or inability to use, this website
• Reliance on information published on this website
• Errors or omissions in published content
• The accuracy, availability or behaviour of third-party platforms, model providers, APIs or messaging services
• Decisions taken on the basis of AI or automation output without adequate human review

Where liability cannot be excluded, it is limited to the fees you paid me in the twelve months before the event giving rise to the claim. Nothing here excludes liability that cannot lawfully be excluded. The fuller position is in the Terms of Service.

I may revise this disclaimer at any time. Changes take effect when posted. The date at the top reflects the current version, and continued use of the website after a change means you accept it.

If something on this website is inaccurate, out of date, or capable of misleading you, please tell me and I will correct it. Contact:

Email: work@amekpoagbe.com

Related documents: Privacy Policy · Terms of Service