The use of Artificial Intelligence (“AI”) in the workplace is likely to become increasingly common as technologies advance and benefits become proven and trusted.
One AI development that looks set to be eagerly embraced by business is the generative pre-trained transformer model designed by developer OpenAI and used in its Chat GPT product (as well as in competing products, such as Google’s Bard and Microsoft’s Bing Chat).
In a nutshell, Chat GPT is a large language model chatbot that is able to generate content based on requirements set by the user. It essentially searches the internet to find information that matches your query and can then produce written content including articles, emails and even social media posts, as well as images and graphics.
While the use of this technology may create huge benefits for your workplace, there are a few key points that you should be aware of before you start:
- If you plan to use Chat GPT in the workplace, you should open a Chat GPT Enterprise account (or its equivalent). The Enterprise account will provide greater speed, data security and privacy than a standard individual account. Yes, it will cost more, but if your employees use Chat GPT from their own account there is a risk that sensitive/confidential information may be leaked and could become available to anyone. If you use the Enterprise level service, any information you input or work product received will remain private to you and will not be used by the model to provide work output for third parties.
- Furthermore, if an employee uses their own Chat GPT account to generate work product, it is the employee who will be the copyright owner of the work product – not the employer. This may create the issue of having to assign or license the right to use the content, or the employee having the ability to share the work product with third parties without requiring your consent to do so. It may be simpler to have copyright clearly belonging to you from the start.
- Another important point is that the content that you receive may not always be correct or suitable! Chat GPT looks for any available information to help it formulate its work product. That means it will not always retrieve accurate information, and in some cases could be basing its work product on false information. Furthermore, the style of the response may not be suited to you or your customers. Chat GPT is not yet able to distinguish between what might be a great communication to an American consumer from one that would work better in a New Zealand context.
If you are thinking about using AI in the workplace, it really is a good idea to put policies in place so that you and your employees understand the parameters and effects. It is an ever-evolving world of technology so you will need to ensure that policies (and employment agreements) are regularly reviewed and updated to meet your needs.
If you require any assistance with your policy drafting and implementation, our employment law team is here to help (and will provide you with a much better-tailored service than Chat GPT!)