4 Ethical Considerations in AI Project Management

Planisware’s founders first met while working in the AI department of a French multinational company in the 1990s. So, when they formed Planisware in 1996, they knew AI and machine learning (ML) would help shape the future of their project portfolio management (PPM) technology. However, for many project management offices (PMOs) and project management (PM) professionals, AI and ML are still relatively new technologies. So, it’s no surprise that many PMOs still have questions about how to use AI ethically.

As environmental, social, and governance (ESG) factors continue to grow in importance, ethical AI use isn’t just about doing what’s right. It’s integral to protecting your business's long-term value. Failure to comply with ESG principles can lead to investment challenges and financial risks, damaging your business’s reputation and overall competitive edge.

PMOs must navigate the ethical considerations in AI project management to maximize the ROI of AI-enhanced PPM. In this article, we’ll explore the associated risks and solutions to confidently embracing AI-enabled PPM solutions with clear guidelines for their ethical use.

What are ethics in AI project management?

“Ethics: a system of accepted beliefs that control behavior, especially such a system based on morals.” — Cambridge Dictionary

Put simply, ethics guide us on “right” and “wrong,” usually setting parameters for principles such as fairness, justice, honesty, respect, and accountability.

At an organizational level, ethics provide a set of standards that people within and outside an organization can use to evaluate its actions and attitudes. So, what are ethics in AI project management? They’re about ensuring that project management (PM) professionals and their teams use AI responsibly and without causing harm. This means:

• Using AI to benefit the organization and the world or industry it operates in
• Ensuring its use doesn’t exploit any person or group, or put them at risk
• Always promoting fairness

These factors resonate with ESG principles, which champion responsible and regenerative practices like sustainability, inclusivity, and compliance. Now, let’s use this insight to explore how ethics applies to AI project management for the PMO, specifically.

4 Ethical Considerations of AI Project Management

The scale of your business’s ethical considerations for AI project management will vary. But, we have identified 4 key areas that all PMOs must consider, no matter the industry they work in or the size of the business. These are necessary to preserve the ROI of AI-enabled PPM practices and tools by complying with ESG standards and avoiding costly reputation damage.

1. The Cultural Impact of AI Project Management

When you introduce an AI-enabled PM tool into your organization, depending on its AI readiness and the maturity of your PMO, you may notice a culture shift. This can be positive — with higher productivity, more time for creativity, and easier collaboration. But it can also have negative consequences if you don’t manage the change properly:

• Fear of job displacement. Because AI and ML can automate so many tasks, some people fear the technology will replace their jobs. While we are a long way from replacing PMOs with AI, the perceived risk can still increase anxiety around job security.
• Resistance, skepticism, and skill gaps. Some team members may lack the skills, knowledge, or confidence to use AI tools effectively. Similarly, some may be reluctant to adopt AI tools based on their personal perceptions of AI. You may need to invest time and budget into providing additional training and support to these team members in particular.
• Changes to the way teams communicate. If your team members find it easier to determine solutions using AI tools, they may become overly reliant on AI to give them the answers they need. This could increase the risk of information gaps, feelings of isolation, and even the loss of important interpersonal skills.

The Solutions

It’s important to remember that AI is still a novel concept to most people. To implement AI in a way that supports a positive, productive culture, you must lead with education, training, and reassurance. This might include:

• Understanding and acknowledging fears about AI and ML. Remind your team that while AI can impact certain job roles, it also creates opportunities and helps PMOs do their jobs more effectively. For example, it can deliver more accurate forecasting, help you identify market trends, and better prioritize your project portfolios based on the most valuable outcomes.
• Opting for user-friendly and intuitive tools that integrate easily with existing systems. When introducing a new tool, you should always implement mandatory training for everyone. And, you should consider providers that can support you during the initial set-up to ensure your teams get the most out of their training time.
• Implementing tools that use intelligent natural language processing (NLP). This will ease communication and ensure accessibility. Remember to simultaneously emphasize the importance of team collaboration to validate AI-generated output and foster opportunities for interpersonal communication.

2. A Lack of Transparency

To work effectively with AI project management tools, we must understand the principles that govern their output. That’s where transparency comes in. You need to know where your data is from, how the algorithm works, and how your tool makes decisions. If tools aren’t transparent, risks include:

• Poor decision-making. ****If an AI-enabled tool lacks “explainability” (the capability to explain its decisions), PMOs risk missing important influences on their projects. As a PM leader, your decision-making and project knowledge stand to benefit significantly from AI’s speed and accuracy. But, an opaque AI tool will make it hard to spot flawed data, pinpoint erroneous outputs, understand and validate its decisions, or even make simple workflow adjustments.
• A lack of trust. Your stakeholders must trust your ability to run projects successfully. Otherwise, you risk constant queries and micromanagement, which can delay projects. This risk increases if they don’t trust the technology you’re using to inform your decisions. Transparent tools make it easier to validate data and provide stakeholders with much-needed visibility.
• Blurred lines of accountability. If PMOs become over-reliant on AI’s recommendations, they may lose a degree of familiarity with and visibility of their projects. This can impact teams’ sense of accountability and ownership regarding project success and failure, making it difficult to determine responsibility.

The Solutions

Transparency is critical to ethical AI project management. So, how do you ensure your AI-enabled PPM tool is transparent, explainable, and trustworthy?

• Select a transparent AI tool that offers users visibility into the decision-making process. This might include choosing tools with open and accessible algorithms, data, and processes. This kind of transparency is particularly valuable for developers who need to understand the intricacies of a system to integrate it with an organization’s wider tech stack.
• Opt for explainable AI (XAI). XAI uses NLP to communicate clearly why its system made particular decisions. This is typically more beneficial to the majority of end users who may not be familiar with the mathematics involved in algorithms.
• Ensure that job descriptions and project rationales include details about project and task ownership. Even AI-generated tasks must include a reference to accountability. This may require regular meetings and communications.

3. Unconscious Biases and Unfairness in Decision-Making

AI-enhanced tools can become biased. This is caused by biases in historical data, human feedback or interaction, and even product development processes. These can lead to harmful and unfair outcomes. Risks include but aren’t limited to:

• Algorithm and operational biases. AI-enhanced PPM tools analyze past project data to help you achieve the best outcomes. But, the algorithm can form biases by overemphasizing one factor as the key to a project’s success. For example, if the algorithm determines a particular person as a project’s key success factor, it may unfairly distribute more tasks to them. Similarly, it may prioritize factors like a project’s budget over its value to the business, quantitative metrics over qualitative skills, and so on. User behavior, preferences, and interactions can further embed these biases in the tool.
• Workplace discrimination and non-compliance. Antidiscrimination laws help protect people against discrimination at work. ****If AI-enabled tools become biased in their decision-making, this can result in prejudices toward protected characteristics such as race, gender, age, socioeconomic factors, geography, and more. For example, if past project data indicates that a male team member has successfully led all projects to date, the tool may begin to favor this particular demographic. This violates antidiscrimination laws and could lead to legal penalties. Discriminatory behaviors, including those delivered by AI, can negatively impact a business's reputation and competitiveness. This could affect your ability to hire new team members, find investors, and more.

The Solutions

To combat biases, it’s vital that you carefully manage the quality of your data and the way you interact with your AI project management solution. To adhere to ethical standards, you must:

• Ensure your data reflects your diverse team. If historical data no longer represents your business, you must update it. Remember, the quality of your data directly impacts your AI project management tool’s outcomes.
• Continuous monitoring. Even the most advanced and trustworthy AI tools develop biases. It’s largely how AI learns. But, ethical AI project management means putting the right tools and tactics in place to mitigate risks and identify harmful biases. You can invest in bias detection technology and practices that proactively identify biases. But, you should also design and implement feedback loops to gain insight from diverse groups. What’s more, you should regularly review and update your data to ensure its quality and accuracy.

4. Data Privacy, Cybersecurity, and Consent

Improving success rates with AI-enabled tools relies on high volumes of high-quality data. But, collecting, storing, accessing, and sharing data comes with risks, especially if that data is sensitive. Risks include:

• Privacy and cybersecurity. No matter what industry you operate in or which AI-enabled tools you use, there will always be data privacy risks and cybersecurity threats. These include data breaches, misuse of personal data, and unauthorized access. This extends to data that hasn’t been anonymized effectively or is retained for long periods without a relevant policy.
• Non-compliance. Your business’s AI tools must comply with data security and privacy regulations. Examples of these include the General Data Protection Regulation (GDPR), which is directly applicable to EU countries, and the California Consumer Privacy Act (CCPA). Governing bodies regularly update their regulations, which vary depending on your country and industry. If they find that your AI tools, or your usage of them, are non-compliant, it can result in legal and financial consequences. This can disrupt your operations and damage your brand perception.

The solution

Managing and mitigating data risks in AI project management underpins its ethical use. So, how do you ensure you and your chosen AI tool handle data safely and securely?

• Data compliance and good data practices. Appoint a team member to take responsibility for compliance with data regulations and follow good practices. Their duties could include championing data storage and sharing best practices, regular data cleansing, and developing a single source of truth for project and portfolio data.
• Working with safe, secure providers. Always check the trustworthiness and quality of the software solutions and other vendors you work with. This includes the AI-enabled PPM tool you opt to use and any third-party providers, integrations, and add-ons.
• Informed consent. It is your responsibility to ensure anyone sharing data with you gives their informed consent for your AI-enabled tool to use it. Employees, customers, and any other users that provide you with their personal data must agree to its use.

How PMOs can work with ethical AI project management in mind

Whether you are an AI novice or you’ve been at the forefront of AI for decades, it’s worth reflecting on its ethical use. If you’re aware of the pitfalls, it’s easier to identify where the risks are in your current systems and usage. Equipped with that knowledge, you can create robust mitigation strategies and embrace the full benefits of AI-enabled PPM.

AI-enhanced PPM tools like Planisware can help PMOs maximize efficiency and boost project success. By implementing Planisware alongside an ethical framework for AI use, PMOs benefit from the following:

• AI-enabled forecasting with easy-access model validation for greater accuracy
• Anomaly detection features that improve data quality
• Intelligent resource allocation and balancing via swarm intelligence (SI)
• Our AI-powered chatbot, Oscar, which enables automated information retrieval
• A single source of truth thanks to a centralized hub of project data