How to Develop Guidelines for Using Artificial Intelligence
Discussions about artificial intelligence (AI) are everywhere — including here. Yet the conversations regarding usage either devolve to the practical and tactical (e.g., how to use the technology) or to the high-minded (e.g., the policies the government should implement and the ethical considerations Big Tech should keep in mind when developing systems). Very little has been written about policies within the world of small to medium business (SMB). We’re going to fix that today.
Access to AI technology has essentially been democratized, and everyone now has the ability to use it. Furthermore, people within your organization are using AI tools — and if they’re not, they should be. Currently, we’re in the “Go-Go-Go / Exploration / Wild West” phase of learning how to leverage the technology. Yet that doesn’t mean it’s too early to begin formulating guidelines for how to use it.
Here are a few questions to consider as you begin the journey for your business.
- What data are we willing to reveal to the AI, and what data are we unwilling to reveal? (i.e., Beware of what you “copy and paste” into the AI text box.)
- What job classifications and functions are best positioned to use AI?
- What are our expectations regarding productivity and output for positions that leverage AI?
- How will individual job descriptions change because of AI?
- What type of feedback mechanisms at the individual and team levels might we provide so we can learn more about how best to use AI within our organization?
- What QC mechanisms must we incorporate into our SOPs and checklists regarding AI-generated content and work product?
- How do we view human error vs. AI-generated errors?
- How do we view an AI-generated error with respect to the AI’s human “supervisor”?
Clients, Vistage members, and other readers of my posts and writings know how heavily I stress the need for organizational core values. So, it should be unsurprising that I recommend using your core values as a primary filter when crafting your organization’s AI usage guidelines. Ask yourself the following questions to guide the process.
- Which of our core values apply to the use of AI?
- Which of our values require new clarifying statements regarding the usage of AI?
- Which of our core values pertains to the level of transparency we must share with our stakeholders regarding our use of AI?
- Do we view AI as a stakeholder? (Even if only in a limited way).
- How does the way we live our values change if we view AI as a stakeholder?
Google’s Seven AI Principles
Google has created seven principles to help guide its development of artificial intelligence applications. I offer these principles as an additional thought-starter for your organization.
Google’s Objectives for AI applications
We will assess AI applications in view of the following objectives. We believe that AI should:
- Be socially beneficial.
- Avoid creating or reinforcing unfair bias.
- Be built and tested for safety.
- Be accountable to people.
- Incorporate privacy design principles.
- Uphold high standards of scientific excellence.
- Be made available for uses that accord with these principles.
Model Guidelines (ChatGPT Speaks!)
This post wouldn’t be complete without:
a) Providing you a set of model guidelines, and
b) Allowing AI-darling-of-the-moment ChatGPT to have a say.
So, I’ve killed two birds with one stone — slightly bending my rule regarding “100% human-generated content” by asking ChatGPT the following prompt:
Create a set of model guidelines for the use of artificial intelligence within small to medium-sized businesses.
The output is quite good, and I hope it will serve as a starting point for your efforts.
Model Guidelines for the Use of Artificial Intelligence Within Small to Medium-Sized Businesses (SMBs) — as written by ChatGPT
- Objective and Purpose: Clearly define the objectives and purposes of implementing AI solutions in the business, ensuring that they align with the company’s values, goals, and vision. AI should be used to improve efficiency, customer satisfaction, and overall business performance.
- Transparency and Communication: Be transparent with employees, customers, and stakeholders about the deployment of AI systems. Communicate the benefits, limitations, and potential risks associated with AI adoption, as well as any changes in roles or responsibilities that may result from AI integration.
- Data Privacy and Security: Ensure that AI systems comply with all relevant data protection regulations and guidelines. Implement strong data security measures to protect sensitive information and respect user privacy. Obtain proper consent when collecting, using, and sharing user data.
- Fairness and Bias: Proactively address biases in AI systems by investing in diverse training data and implementing regular audits to identify and mitigate any unfair or biased behavior. Foster an inclusive environment that welcomes input from employees and stakeholders from diverse backgrounds.
- Accountability and Responsibility: Assign responsibility for AI systems and their outcomes to appropriate team members. Establish a system for monitoring and evaluating the performance of AI applications, and ensure that decision-makers are accountable for the consequences of their use.
- Employee Training and Support: Provide training and resources to help employees adapt to and leverage AI systems effectively. Encourage ongoing learning and development to build a workforce that is skilled in working with AI technologies.
- Safety and Reliability: Prioritize the development and deployment of safe and reliable AI systems. Test AI applications rigorously before deployment and continuously monitor their performance to identify and resolve potential issues promptly.
- Ethical Considerations: Ensure that AI systems align with ethical guidelines and societal values. Consult with relevant stakeholders, including employees, customers, and industry experts, to develop and maintain a strong ethical framework for AI use.
- Collaboration and Partnerships: Collaborate with other businesses, industry experts, and regulatory bodies to share best practices, insights, and challenges related to AI adoption. Build partnerships to drive innovation and improve the responsible use of AI across the business ecosystem.
- Continuous Improvement: Embrace a culture of continuous improvement and adaptation. Regularly assess the impact of AI on the business and refine strategies to optimize benefits while minimizing risks. Stay up-to-date with AI advancements and regulatory changes to ensure ongoing compliance and competitiveness.
Final (Human) Thoughts on AI Use by SMBs
Artificial Intelligence is simultaneously the shiny object of the moment as well as a fundamentally disruptive technology. It will radically transform how we work and reshape our economy in ways we have yet to understand. Therefore, it’s incumbent upon leaders of organizations everywhere to be proactive regarding using AI in their organizations. I respectfully suggest a combination of both restraint and foresight when doing so. As always, I’m here to assist and welcome a conversation with you. Drop me a line — or draft a letter to me using ChatGPT — and we’ll figure things out together.