Ho Wai CHENGProgram Manager| BankHong Kong, Hong Kong
AI tools like ChatGPT and other public LLMs are quickly becoming part of everyday project work. Teams use them to draft communications, summarize documents, analyze data, and speed up routine tasks. From a productivity perspective, the value is clear. From a project governance and risk perspective, it’s more complex.
In regulated or risk‑sensitive environments, even well‑intentioned AI use can introduce concerns around data privacy, compliance, auditability, intellectual property, and decision accountability—often without project managers being fully aware. Shadow AI usage is becoming a real challenge, especially as teams adopt tools faster than policies evolve.
AI can absolutely enhance delivery, decision‑making, and efficiency—but only if it’s used with the right guardrails, transparency, and leadership oversight. As project professionals, we’re accountable not only for speed and outcomes, but also for trust, control, and responsible execution.
Curious to learn from the community:
How is AI currently being used on your projects?
Are public AI tools formally governed, partially restricted, or unmanaged?
Where do you see the biggest benefit—and the biggest risk—today?
Looking forward to hearing real‑world experiences and perspectives.
Saving Changes...
Sort By:
Luis BrancoCEO| Business Insight, Consultores de Gestão, LdªCarcavelos, Lisboa, Portugal
Excellent framing of the topic. The tension between productivity and governance is real, but the critical issue is not the technology, it is the design of the decision system.
Responding to the questions directly:
How is AI being used on projects? Primarily as a cognitive assistant. Drafting communications, summarising information, exploratory analysis and clarification of options. In many cases, it already influences project reasoning, even when it is described merely as “support”.
Are public AI tools governed, partially restricted, or unmanaged? In practice, partially restricted in discourse and weakly governed in operation. There are usually generic IT and information security policies, but rarely explicit guidance on usage boundaries, decision accountability, traceability, or conscious human validation.
Biggest benefit today? Cognitive leverage. AI improves the quality of early thinking and frees time for human judgment, leadership and systemic integration, when properly framed.
Biggest risk today? Implicit delegation of judgment without organisational awareness. When AI starts shaping decisions, narratives or priorities without responsibility, assumptions and criteria being clearly anchored in humans.
Perhaps the core challenge is this. Many organisations still treat AI as an operational tool, when in reality it already acts as a decision amplifier. That requires AI use to be integrated into project governance from the outset, with clarity on what is allowed, what requires explicit human validation, and what should not be delegated at all.
Well-designed governance does not slow teams down. It creates trust, coherence and sustainable execution.
...
1 reply by Ho Wai CHENG
Feb 02, 2026 10:29 AM
Ho Wai CHENG
...
Thank you for the clear and insightful response. Your framing of AI as a decision amplifier rather than merely an operational tool is particularly helpful and resonates strongly with the realities many organisations are facing.
I agree that the most significant risk today is the implicit delegation of judgment without sufficient organisational awareness. As you noted, AI already influences reasoning and narratives, even when positioned as “support,” which makes explicit governance and human validation essential rather than optional.
Your observation that governance is often weakly defined in practice—relying on generic IT or security policies rather than clear decision boundaries and accountability—is especially relevant in regulated and risk‑sensitive environments. Treating AI use as part of the project decision system from the outset feels like a necessary shift.
I also appreciate your point that well‑designed governance does not slow teams down, but instead builds trust and enables sustainable execution. That balance between cognitive leverage and accountable decision‑making is exactly where I see the conversation needing to move.
Thank you for articulating this so clearly.
Saving Changes...
Sergio Luis ConteHelping to create solutions for everyone| Worldwide based OrganizationsBuenos Aires, Argentina
You are talking about generative AI which is just a subset of AI. What you are asking is not related to AI in general. Returning to generative AI what most of the people and organizations are not aware is about the key component called Responsible AI. This is the key ingredient do not fail and do not have serious consequences.
...
1 reply by Ho Wai CHENG
Feb 01, 2026 11:29 AM
Ho Wai CHENG
...
Thank you for the clarification—you’re absolutely right that generative AI is a subset of AI. I also agree with your point on Responsible AI being a critical component. From a project and delivery perspective, this is precisely where the conversation becomes relevant: teams are already using generative AI tools, often ahead of formal policies. That creates both opportunity and risk. Embedding Responsible AI principles—such as transparency, data protection, and human oversight—into project governance is becoming essential to avoid unintended consequences.
So while the broader AI landscape is much larger, the questions raised here are meant to address a very practical challenge for project professionals: how to enable productivity gains from generative AI while ensuring responsible, compliant, and well‑governed use.
Appreciate your perspective and would be interested in how to operationalize Responsible AI.
Saving Changes...
Ho Wai CHENGProgram Manager| BankHong Kong, Hong Kong
Feb 01, 2026 10:16 AM
Replying to Sergio Luis Conte
...
You are talking about generative AI which is just a subset of AI. What you are asking is not related to AI in general. Returning to generative AI what most of the people and organizations are not aware is about the key component called Responsible AI. This is the key ingredient do not fail and do not have serious consequences.
Thank you for the clarification—you’re absolutely right that generative AI is a subset of AI. I also agree with your point on Responsible AI being a critical component. From a project and delivery perspective, this is precisely where the conversation becomes relevant: teams are already using generative AI tools, often ahead of formal policies. That creates both opportunity and risk. Embedding Responsible AI principles—such as transparency, data protection, and human oversight—into project governance is becoming essential to avoid unintended consequences.
So while the broader AI landscape is much larger, the questions raised here are meant to address a very practical challenge for project professionals: how to enable productivity gains from generative AI while ensuring responsible, compliant, and well‑governed use.
Appreciate your perspective and would be interested in how to operationalize Responsible AI.
...
1 reply by Sergio Luis Conte
Feb 02, 2026 8:35 AM
Sergio Luis Conte
...
You are welcome. Responsible AI is a matter to define process, methods, skills to be engaged. And if generative Ai will be used in the whole organization then it is a matter of strategy. Roles that belongs to Responsible Ai belongs to areas like legal, linguistic, diversity and inclusion, etc, etc
Saving Changes...
Sergio Luis ConteHelping to create solutions for everyone| Worldwide based OrganizationsBuenos Aires, Argentina
Feb 01, 2026 11:29 AM
Replying to Ho Wai CHENG
...
Thank you for the clarification—you’re absolutely right that generative AI is a subset of AI. I also agree with your point on Responsible AI being a critical component. From a project and delivery perspective, this is precisely where the conversation becomes relevant: teams are already using generative AI tools, often ahead of formal policies. That creates both opportunity and risk. Embedding Responsible AI principles—such as transparency, data protection, and human oversight—into project governance is becoming essential to avoid unintended consequences.
So while the broader AI landscape is much larger, the questions raised here are meant to address a very practical challenge for project professionals: how to enable productivity gains from generative AI while ensuring responsible, compliant, and well‑governed use.
Appreciate your perspective and would be interested in how to operationalize Responsible AI.
You are welcome. Responsible AI is a matter to define process, methods, skills to be engaged. And if generative Ai will be used in the whole organization then it is a matter of strategy. Roles that belongs to Responsible Ai belongs to areas like legal, linguistic, diversity and inclusion, etc, etc
...
1 reply by Ho Wai CHENG
Feb 02, 2026 10:33 AM
Ho Wai CHENG
...
Thank you for your perspective. Your point that organisation‑wide use of generative AI becomes a strategic matter is especially important. When AI is embedded broadly, responsibility cannot sit with delivery teams alone—it requires cross‑functional ownership across areas such as legal, risk, ethics, and inclusion, working in alignment.
From a project and delivery standpoint, this reinforces the need to translate those enterprise‑level principles into practical guidance: clear usage boundaries, explicit decision accountability, and skills that enable teams to apply AI responsibly in day‑to‑day work.
Saving Changes...
Ho Wai CHENGProgram Manager| BankHong Kong, Hong Kong
Jan 31, 2026 10:01 AM
Replying to Luis Branco
...
Excellent framing of the topic. The tension between productivity and governance is real, but the critical issue is not the technology, it is the design of the decision system.
Responding to the questions directly:
How is AI being used on projects? Primarily as a cognitive assistant. Drafting communications, summarising information, exploratory analysis and clarification of options. In many cases, it already influences project reasoning, even when it is described merely as “support”.
Are public AI tools governed, partially restricted, or unmanaged? In practice, partially restricted in discourse and weakly governed in operation. There are usually generic IT and information security policies, but rarely explicit guidance on usage boundaries, decision accountability, traceability, or conscious human validation.
Biggest benefit today? Cognitive leverage. AI improves the quality of early thinking and frees time for human judgment, leadership and systemic integration, when properly framed.
Biggest risk today? Implicit delegation of judgment without organisational awareness. When AI starts shaping decisions, narratives or priorities without responsibility, assumptions and criteria being clearly anchored in humans.
Perhaps the core challenge is this. Many organisations still treat AI as an operational tool, when in reality it already acts as a decision amplifier. That requires AI use to be integrated into project governance from the outset, with clarity on what is allowed, what requires explicit human validation, and what should not be delegated at all.
Well-designed governance does not slow teams down. It creates trust, coherence and sustainable execution.
Thank you for the clear and insightful response. Your framing of AI as a decision amplifier rather than merely an operational tool is particularly helpful and resonates strongly with the realities many organisations are facing.
I agree that the most significant risk today is the implicit delegation of judgment without sufficient organisational awareness. As you noted, AI already influences reasoning and narratives, even when positioned as “support,” which makes explicit governance and human validation essential rather than optional.
Your observation that governance is often weakly defined in practice—relying on generic IT or security policies rather than clear decision boundaries and accountability—is especially relevant in regulated and risk‑sensitive environments. Treating AI use as part of the project decision system from the outset feels like a necessary shift.
I also appreciate your point that well‑designed governance does not slow teams down, but instead builds trust and enables sustainable execution. That balance between cognitive leverage and accountable decision‑making is exactly where I see the conversation needing to move.
Thank you for articulating this so clearly. Saving Changes...
Ho Wai CHENGProgram Manager| BankHong Kong, Hong Kong
Feb 02, 2026 8:35 AM
Replying to Sergio Luis Conte
...
You are welcome. Responsible AI is a matter to define process, methods, skills to be engaged. And if generative Ai will be used in the whole organization then it is a matter of strategy. Roles that belongs to Responsible Ai belongs to areas like legal, linguistic, diversity and inclusion, etc, etc
Thank you for your perspective. Your point that organisation‑wide use of generative AI becomes a strategic matter is especially important. When AI is embedded broadly, responsibility cannot sit with delivery teams alone—it requires cross‑functional ownership across areas such as legal, risk, ethics, and inclusion, working in alignment.
From a project and delivery standpoint, this reinforces the need to translate those enterprise‑level principles into practical guidance: clear usage boundaries, explicit decision accountability, and skills that enable teams to apply AI responsibly in day‑to‑day work. Saving Changes...
