Strong Contents policy on the use of AI tools for employees and subcontractors

This policy applies to all content produced by Strong Contents, whether for internal or external use.

Introduction

Strong Content’s core business is to deliver the highest quality content. Now more than ever, we need human validation of machine-generated content. Not only to combat the spread of misinformation and fake news, but also to maintain human creativity and skills. AI-generated content has not yet reached a level where it can generate high-quality content and texts that meet our standards as a content agency.

Strong Contents recognises the potential benefits of using AI to improve efficiency and productivity in the workplace. But we also recognise the importance of using AI responsibly and ethically, especially when it comes to generating content. Of course, we need to adapt and master AI tools to get better. We believe that the value of a quality content creator/writer/editor/translator who also knows how to utilise AI content will increase enormously in the future.

Purpose of this policy

Our policy aims to provide guidelines for the responsible use of AI-generated content, emphasising the need for proofreading, editing, fact-checking and using AI-generated content as a starting point, not as the finished product.

As a content agency that delivers on its promise of high quality content to its clients, it is of utmost importance that everything we publish – whether for our own website, blog posts, social media and other channels, or for clients – is produced by a human.

Tools like ChatGPT, DeepL, Byword.AI, Canva AI and others are just that – tools. They can help us structure data and information, but they cannot replace us in researching, writing and editing that data and information.

Strong Content’s policy is that ChatGPT should not produce entire works such as blog posts, social media posts, newsletters, interviews and other content formats intended for external publication.

We are writers and content experts, and we need to keep honing our skills, doing our own research and writing our own texts.

Use ChatGPT as an assistant who can provide ideas, help structure a blog post if you’re stuck, or check your post for spelling errors. The same goes for translations – no online tool is currently as good a translator as a real translator. You can use translation tools such as DeepL or Google Translate to make a first rough translation, but you must always check, read and edit to improve the translation.

Exceptions

There may be a joint decision (internally and with the client) to generate content using ChatGPT, if it is to bulk produce SEO articles whose sole purpose is to generate website traffic. Then a copy of all prompts must be saved in a common document, to learn and see if we can create better articles using better prompts. These articles must always be edited and fact-checked before publication and it must always be clear that ChatGPT was part or all of the creation of this content.

Requirements for AI-generated content:

Proofreading: All AI-generated content must be proofread and checked by a human before being published or shared. This includes checking for spelling, grammar and factual errors.
Editing: AI-generated content must be edited to ensure it is well-written, coherent and engaging. This includes ensuring that the content is structured in a logical way and that it is appropriate for the intended reader.
Fact-checking: AI-generated content must be fact-checked to ensure that all information is accurate and up-to-date. This includes verifying sources, checking statistics, and ensuring that any claims made in the content are supported by evidence.
Starting point, not the finished product: AI-generated content should be seen as a starting point, not a finished product. AI can be a valuable tool for generating content, but it cannot replace the creativity and critical thinking of human writers and editors.

Guidelines for the responsible use of AI-generated content:

Transparency: All AI-generated content must be clearly labelled as such, and the use of AI to generate content should be transparent to employees and customers
Data privacy: We must ensure that any personal or sensitive data used to train AI models is handled with appropriate care and that any AI-generated content containing such data is handled in accordance with data protection laws.
Fairness: We must ensure that AI-generated content does not discriminate against any individual based on their protected characteristics, such as ethnic origin, gender, age or disability.
Accountability: Our company must take responsibility for any damage caused by AI-generated content, and we must have insurance cover to protect us against any legal claims.

Warning! Do not share proprietary data with ChatGPT

As an AI language model, ChatGPT is designed to process and generate language based on the input it receives. ChatGPT is a sophisticated tool that can provide useful insights and answers, but it is not equipped to handle sensitive information.

As ChatGPT is a digital tool that can potentially be accessed by others, it is not appropriate to share protected and confidential information with the tool. There is a risk that the information could be exposed or misused, either through a security breach or by unintended parties gaining access to it. ChatGPT saves all information and can reuse it in its responses to other users, thus sharing sensitive information.

In addition, ChatGPT is not a legal entity and is not bound by the same confidentiality agreements or legal protections as human employees or contractors. As such, it may not be able to guarantee the same level of discretion and confidentiality that a trusted human partner could provide.

In summary, companies should exercise caution when sharing proprietary and confidential information with ChatGPT, as it may pose a security risk and potentially compromise sensitive data.

Some parts of this policy are translated and adapted from this template: https://trainual.com/template/chatgpt-policy