Discussion Board 1.3: Think & Share💡From Prompting to Partnership: When Does AI Become a Digital Teammate?

When I first started out, I was really just using AI like a search engine—nothing fancy. But that's changed a lot. These days I'm using it to build training materials, automate the routine reporting, update templates... it's honestly become like a digital teammate for me at this point, handling the repetitive stuff so I don't have to.
That said, it's not a "set it and forget it" tool. I still need to give it clear, specific instructions, check the output carefully, and tweak things until they actually fit what the business needs. AI doesn't replace judgment—it just takes some of the grunt work off my plate.
As a leader, though, my biggest worry is still data security. We're simply not putting sensitive data anywhere near AI tools right now—not worth the risk.
And honestly, I don't think most organizations are actually ready for this yet. Two things stand out to me:
People aren't there yet. A lot of users either haven't picked up AI skills at all, or they're stuck just using it as a basic chatbot—they haven't explored what else it can actually do for them.
The cost catches people off guard. I've seen companies jump in, cut staff assuming AI would cover the gap, and then have to walk it back because the real cost of running AI properly ended up way higher than they budgeted for.

My interaction with AI has clearly evolved from information retrieval to collaboration. A good example is when I started developing an algorithmic trading system. Initially, I used AI to answer technical questions about Python and trading strategies. Over time, the interaction became iterative: I would propose a strategy, test it, bring back the results, challenge the assumptions and redesign the model with AI. At that point, AI was no longer simply providing answers but participating in the thinking process.

I see an even greater opportunity in corporate banking and credit analysis. Instead of reviewing credit risk only at predefined intervals, an AI digital teammate could continuously analyse multiple sources, identify emerging red or green flags and trigger deeper reviews when circumstances change. It could also dynamically reassess recommended banking exposure.

Human judgement would remain essential. AI can identify patterns and signals, but understanding management quality, context or an unconventional business model still requires experience and judgement. As highlighted in the module, when intelligence becomes cheaper, human judgement becomes more valuable. 

The main leadership concern is overreliance. Employees may gradually accept AI recommendations without challenging them. Leaders must clearly define where AI can act, where humans intervene and who remains accountable.

I do not believe most organisations are ready yet. Many are simply adding AI to old workflows rather than redesigning how work is done. The real shift will come when organisations build processes around collaboration between humans and digital teammates.

1. Evolution of Interaction

Over the years, my interaction with AI has evolved from simply using it as a search engine or looking for direct answers among a long list of search results. Today, AI has become part of my daily workflow, whether for quick searches, research, proofreading documents, or learning new topics.

The biggest change has been moving from using AI as an information retrieval tool to treating it as a collaborative assistant that helps me think, refine ideas, and work more efficiently. While fact-checking is still necessary because AI is not always correct, it has significantly reduced the time required to complete many tasks.

2. Human + AI Workflow Thinking

What role the AI would play

AI is becoming increasingly advanced, with larger models and more parameters enabling it to process vast amounts of information and generate meaningful outputs. Although we are still far from achieving true AGI, AI has already become an additional “brain” that allows humans to accomplish more in less time by supporting research, analysis, and knowledge discovery.

Where human judgment would remain essential

Human judgment remains critical because the data used to train AI may not always be accurate or complete. AI systems can produce false positives or misleading information, and their outputs still require human evaluation and validation. The responsibility for assessing accuracy, context, and appropriateness ultimately remains with people.

How the collaboration could improve outcomes

Human-AI collaboration can significantly improve productivity by producing outcomes faster and allowing people to focus on creativity, strategic thinking, and innovation. AI can quickly analyse information, identify patterns, and draw insights from historical data, enabling humans to spend more time on forward-thinking and decision-making.

3. Leadership Implications

With rapid technological advancements comes the challenge of continuously learning new technologies. Technology is evolving exponentially, but our ability to learn and adapt does not naturally progress at the same pace. AI can help bridge this gap, but human involvement remains essential.

As AI becomes more integrated into organisations, leaders need to provide sufficient training and guidance to help employees develop the skills required to work effectively with these technologies. Learning should go beyond simply watching training videos; employees need opportunities to apply these skills in practical situations. Ultimately, the best way to adapt to emerging technologies is through hands-on experience and continuous practice.

4. Reflection on AI Maturity

Since the introduction of ChatGPT, organisations have started to recognise the potential and broad applications of AI. However, concerns around privacy, security, and governance are still catching up with the pace of technological development.

Because of these concerns, many leaders remain cautious and often take a risk-averse approach when adopting AI within their workflows. While there is growing interest in collaborative AI systems, I believe many organisations are still in the early stages of building the policies, capabilities, and confidence needed to fully embrace AI as a true workplace collaborator.

1. Evolution of Interaction

Describe a situation where your interaction with AI evolved beyond simple information retrieval into something more iterative or collaborative.

I experienced this while preparing a presentation for a "show and tell" session aimed at leaders who weren't particularly tech-savvy. Rather than simply asking AI for content, I used it iteratively - feeding it my raw ideas, having it help structure the narrative logically, and going back and forth to rephrase technical concepts into language that non-technical leaders could easily grasp. Each round of feedback refined the output further. The session was well received, and my team got a lot of positive feedback afterward.

What changed in the interaction dynamic?

The interaction shifted from a one-way query-response exchange to a genuine back-and-forth refinement process. Instead of treating AI as a source of answers, I treated it more like a co-editor - someone I could iterate with until the output matched exactly what I needed. That shift in dynamic is what shaped the final result so effectively.

2. Human + AI Workflow Thinking

Identify one workflow, decision process, or professional activity where you believe AI could function as a meaningful "digital teammate."

I believe AI could serve as a digital teammate in resolving repetitive support issues. A large portion of support queries tend to repeat themselves - questions like "what access do I need to request," "why can't I view a particular dashboard," "why are my numbers different from someone else's," or "how do I use this feature." AI could handle these automatically by referencing our ticketing logs and knowledge repository, escalating to a human only when it can't find a reliable answer.

What role the AI would play:
AI would act as an enabler - resolving repetitive, low-complexity issues and cutting through the noise so that human effort is directed toward genuinely complex or novel problems.

Where human judgment would remain essential:
Human judgment becomes essential for issues that fall outside the knowledge repository, or in situations that require a high-stakes decision needing formal approval or accountability.

How the collaboration could improve outcomes:
This collaboration would free up human bandwidth, broaden the team's capacity for higher-value thinking, and significantly reduce resolution time for both simple and complex issues.

3. Leadership Implications

From a leadership perspective, what opportunities or concerns emerge when employees begin interacting with AI systems as collaborators rather than tools?

When leaders genuinely understand how and why AI is being used, the implications tend to be positive. That said, both leadership and employees need to adapt to the shift AI brings. This requires two layers of safeguards: first, individuals need to stay within the guardrails set by the organisation and remain accountable for only sharing information they're authorised to share; second, the organisation's compliance policies act as a backstop, catching anything that slips through individual judgment.

Trust also has to be built on both sides - leadership needs to trust employees to use AI responsibly, and employees need to trust that AI adoption is being done thoughtfully, not just for optics. AI shouldn't be adopted simply because it's trendy; organisational readiness needs to be properly assessed first, and so does team dynamics - whether the team as a whole is genuinely ready to adopt and move forward together. Where there's hesitation or resistance, targeted training sessions led by someone credible and trusted within the team can help demonstrate AI's real, tangible benefits.

4. Reflection on AI Maturity

Do you believe most organisations today are psychologically and operationally ready for collaborative AI systems? Why or why not?

Most legacy systems were never designed with AI in mind, which creates a structural gap. Some organisations adopt AI purely to appear innovative or to show they're "investing in the future," without a real strategy behind it. Organisations that genuinely understand AI's practical value, however, are actively redesigning their systems and processes to work with AI from the ground up — treating AI-readiness as a core design principle for anything new they build.

I've seen this firsthand. When we implemented an AI support chatbot, "Ask Dynamics," we trained it on our internal knowledge repository, gave it clear starter prompts, and set strict boundaries to prevent hallucination and ensure proper escalation. The result was a 60% reduction in support queries reaching our developers - a clear example of what's possible when readiness and implementation are done right.

My use of AI has evolved from using it as a search engine to command-collaborate-improve-repeat. Months ago while testing the waters, I took a  picture of 10 receipts and asked ChatGPT to tabulate invoice numbers, vat numbers, and individual costs of each receipt.

My command to ChatGPT: tabulate invoice numbers, vat numbers, and individual costs  for 10 receipts

ChatGPT: receipt 2 is blurred but invoice number may be 12345. The invoice number is taken from the upper right hand corner.

Collaboration: here’s a new picture, with the invoice numbers clicked in green (ChatGPT got it wrong the first time), vat numbers (circled in red), and individual costs (circled in blue).

ChatGPT tabulated all in less than a minute  what I would normally take some 15 mins to do. It also  gave  some nice suggestions.

While I was happy how the task was done so quickly (saving me the time and eye strain :), I did worry about data security ( would my receipts be recorded somewhere forever?) Ethically, was I cutting corners? I don’t think I was but it’s a question I asked myself. I think many leaders probably don’t know how to reply, too. My aim is to educate myself so I can make good decisions on the use of AI or help a fellow leader to do.

From a personal expense report, I am now considering how to best make use of AI to spot where workflows may have fallen for some 10000 test records for my company.

Reflection on Human–AI Collaboration

1. Evolution of Interaction

My interaction with AI began mainly with information retrieval, summarising documents, and improving written communication. Over time, it became more iterative and collaborative.

For example, I used AI to analyse patient experience feedback by theme and quarter. Instead of accepting the first output, I refined the prompts, added operational context, compared results across models, and challenged assumptions. The AI became more of a thinking partner than a search tool.

This experience also showed me that AI outputs can sound convincing even when incomplete. Human validation remained essential, especially when interpreting findings and deciding what actions were appropriate.

2. Human + AI Workflow Thinking

One area where AI could act as a digital teammate is patient safety incident review.

AI could retrieve relevant documents, summarise the sequence of events, highlight missing information, identify similar past incidents, and flag cases that may require escalation.

Human judgment would still be essential in assessing clinical seriousness, understanding context, engaging staff and patients, and deciding follow-up actions. These decisions often involve ambiguity, ethics, accountability, and sensitivity.

This collaboration could reduce administrative work, improve consistency, speed up triage, and allow reviewers to focus more on learning and system improvement.

3. Leadership Implications

Collaborative AI offers opportunities to improve productivity, creativity, and access to specialist capabilities such as data analysis, writing, and workflow design.

However, leaders must address trust and accountability. Employees may over-rely on confident AI outputs without checking them. Accountability should remain with the human decision-maker, particularly in healthcare.

AI may also change team dynamics. Early adopters may become more productive, while others may feel threatened or left behind. Leaders need to provide training, psychological safety, clear governance, and opportunities for staff to learn.

4. Reflection on AI Maturity

I do not believe most organisations are fully ready for collaborative AI.

Psychologically, some employees see AI either as a threat or as a system that can be trusted too easily. Operationally, many organisations still have fragmented data, unclear ownership, inconsistent processes, and limited governance.

I have also observed that organisations may be enthusiastic about AI pilots but less prepared for production issues such as integration, cybersecurity, maintenance, user support, and outcome measurement.

However, organisations do not need to wait until everything is perfect. They can begin with lower-risk workflows, involve users, maintain human oversight, and build capability gradually.

Yes can't agree more on the interaction dynamic, back then it was search engine, and today it is AI as search engine, but wait, its more than that. 

I believe AI can be a digital team mates - especially in my day to day professional working environment in tech, take 1 example to prepare presentation material, where AI can help to create a structure based on our objectives (prompts) and data points that we gave to AI. The human judgement would be still exists, to review the slides, verify the truth (whether AI is hallucinating or not), and do the final content review. There will be less time to finalize a deck.

From leadership perspective:

• Opportunity: Productivity increase, expected more outcome per headcount.
• Concern/Challenges: Everyone need to adapt with a new tools. The AI pace today is very different with what we typically had previously, eg. search engine. The AI pace is way way too fast for organization to adapt. The workforce mindset is something need to be carefully taken care of, eg. afraid for job-cut, job relocation, etc

AI Maturity? It's challenging, people already get used to what they have today from technology perspective. Based on my experience, people don't like change (tools, way of working). So that's why I think Change Management division is in place. Because it is the most challenging part.

Evolution of Interaction. My journey with AI began about a year ago using ChatGPT and Gemini, primarily as cost-effective, efficient tools to summarize and validate complex tax laws. By the end of last year, my usage shifted toward drafting corporate tax policies, generating compliance documentation, and creating executive presentations. This evolution was driven by leadership’s increasing demands for faster, high-quality deliverables under tight deadlines.

Human + AI Workflow Thinking.  A primary use case has been developing localized transfer pricing documentation. I leverage AI to draft core content based on OECD guidelines, using it to write extensive sections on industry benchmarks and market segments. I then input our company’s specific functional profile, and AI seamlessly integrates the two. However, human oversight remains critical; AI occasionally cites incorrect tax codes or assumes inaccurate tax positions, making rigorous proofreading non-negotiable.

Leadership Implications & Team Dynamics. While AI enhances productivity, I am increasingly concerned about over-reliance. A critical risk is that heavy dependence AI may erode critical thinking and creativity over time. Furthermore, if team members accept AI outputs without thorough validation, inaccuracies will slip through. This requires a shift in team dynamics, moving our focus heavily toward quality assurance and risk management.

Reflection on AI Maturity.  AI adoption is an inevitable industry shift driven by the pursuit of efficiency, automation, and cost reduction—which will naturally result in role redundancies. In my last two roles, C-suite executives mandated KPIs for AI deployment, expecting tangible financial savings. However, a significant "expectation gap" remains between what AI can realistically achieve today and what leadership expects it to deliver.  

1. Evolution of Interaction

My interaction with AI shifted from lookup to iteration through my own consulting work. Early on I'd ask for a training outline or a deck and rebuild the context every single time. Over time I moved to persistent "skills": standing templates for quotations, invoices, training outlines, and recurring stakeholder reporting decks that already encode my formatting rules, brand voice, and structural conventions. The dynamic changed from "give me an answer" to "build on what we already agreed." I'm no longer briefing from zero, I'm supervising and refining.

2. Human + AI Workflow Thinking

The clearest candidate is a course verification workflow I run for regulatory training submissions: mapping a training outline into structured learning outcomes, cognitive levels, and outcome categories required by the framework. AI can produce a fast, consistent first-pass mapping and flag structural gaps (an outcome with no matching activity, a category mismatch) that are easy to miss under deadline pressure. Human judgment stays essential on two fronts: confirming the pedagogical intent is actually correct, not just correctly formatted, and owning the regulatory sign-off. A wrong claim in a compliance submission is a problem I carry, not the AI. Done well, this cuts drafting time and reduces the "drift" that creeps in whenever content passes through multiple hand-offs.

3. Leadership Implications

Once people treat AI as a collaborator, trust becomes "when do I need to double-check this," not "does it work," and that has to be trained, not assumed. Accountability doesn't move: "the AI wrote it" is not a defence for a non-compliant submission, and leaders need to say that out loud before someone learns it the hard way. Creativity risks converging: if everyone's first draft comes from the same model with similar prompts, output starts looking alike unless people are pushed to diverge from the default. And organisational readiness is the real gap. Most places I work with have individuals experimenting well one-on-one, with no shared standards, no review gate before AI-drafted content goes external, and no institutional memory of what "good" looks like.

4. Reflection on AI Maturity

No, not operationally, and mostly not psychologically either. Across the training and compliance-heavy clients I work with, individual competence is outpacing organisational systems: people use AI well personally, but there's no shared prompt library, no review gate before AI-drafted content reaches a regulator or a board, and no clarity on who owns an error. That's a governance gap, not a technology one. Readiness will show up first in compliance-heavy sectors forced to formalise it, not in the sectors most excited about AI.

1. Evolution of Interaction
I used AI mainly to retrieve information, summarise policies, and clarify regulatory requirements. Over time, my interaction became more collaborative. I now use AI to refine audit reports, brainstorm control improvements, develop risk assessment frameworks, and review communication before engaging stakeholders. Instead of simply providing answers, AI has become an iterative partner that helps improve the quality and structure of my work while I validate the final output.

2. Human + AI Workflow Thinking
In operations audit and compliance, AI could serve as a digital teammate during the audit planning and reporting process. It can analyse large volumes of audit data, identify trends, highlight potential control gaps, and generate draft reports or recommendations. Human judgment remains essential to assess business context, determine the significance of findings, evaluate regulatory implications, and make risk-based decisions. This collaboration would improve efficiency, consistency, and allow auditors to focus on higher value analysis and stakeholder engagement.

3. Leadership Implications
As employees begin working with AI as collaborators, leaders have an opportunity to improve productivity, decision making, and innovation. However, they must also address concerns around trust, accountability, and responsible AI usage. Employees need clear governance, training, and defined ownership so that AI supports decisions rather than replaces professional judgment. Strong leadership is critical to ensure AI is used ethically and effectively.

4. Reflection on AI Maturity
I believe most organisations are still in the early stages of AI maturity. While many have adopted AI tools, not all have established the governance, data quality, or employee capability required for effective human-AI collaboration. Organisations that invest in AI literacy, governance frameworks, and change management will be better positioned to realise the full value of collaborative AI while managing associated risks.

Evolution of Interaction

I first used AI mainly to rewrite website content. Later, I started using it to support GEO by identifying missing questions, improving the structure, and considering how people might search. It became more collaborative because the work involved several rounds of testing and refinement. AI helped me improve the content and think through different options, while I still made the final decisions.

Human + AI Workflow Thinking

AI can act as a digital teammate in competitive intelligence. It can review public information, summarize developments, and highlight relevant changes. It can also be trained on the business, making the output more useful. Human judgment is still needed to verify information, understand context, and determine what is actually important.

Leadership Implications

The main opportunities are speed, creativity, and better use of limited resources. The risks include inaccuracy, overreliance, unclear accountability, and data security. I have seen some colleagues rely too heavily on AI-generated content without reviewing or improving it. This can result in work that feels generic or disconnected from the actual audience. Leaders need to make it clear that AI should help people think and work better, rather than replace their judgment. Individuals still need to review, refine, verify, and take responsibility for the final work. Companies also need approved tools, practical training, and simpler policies that employees will actually read.

Reflection on AI Maturity

I think the answer is yes and no. At a recent congress, I saw many organizations already using AI agents and automated workflows. However, in most organizations, especially in regulated industries, adoption is still slowed by IT, legal, compliance, and regulatory requirements. The technology is moving quickly, but many organizations are still catching up.

1. Evolution of Interaction

EvolIn my industry of cloud SaaS products, AI has really changed the game! We've moved from using traditional machine learning and NLP for things like data classification and search suggestions to the exciting world of Generative AI and autonomous Agentic AI systems. At first, SaaS platforms were mainly used to keep track of information, using NLP and machine learning for simple tasks like sorting data, suggesting searches, and doing basic analytics. When I first got into Generative AI, I was mostly working with search queries, summarizing text, and writing. But then, it quickly became about doing things, using advanced models to handle tricky data, dive deep into research, and do what-if analyses. Now, the market is heading towards Agentic-as-a-Service (AaaS), where autonomous, goal-driven systems can run complex business workflows with little help from us. To keep up with this, I'm building and using special AI agents to automate lots of complicated tasks. This includes using low-code agent studios and orchestration tools to manage repeated thinking, create data visualizations, and get ready for presentations and infographics.

2. Human + AI Workflow Thinking

As a technologist, conducting in-depth topical research on a daily basis is crucial. AI models can effectively perform this task as a preliminary step, although they often tend to incorporate outdated or irrelevant sources, necessitating human oversight for curation. Additionally, since commercial AI models can be biased toward confirming user prompts, it is essential to critically examine and challenge their outputs. Finally, while AI has become highly proficient in synthesizing visualisations of data and information, I primarily utilise it to generate the initial draft of presentations, which then requires rigorous review and multiple iterations of human refinement. These modes of collaboration have saved me hundreds of hours in terms of repetitive work and detailed research work, enabling me to achieve higher productivity. However, to establish an effective workflow, there is a requirement for appropriate human-AI orchestration tools. Most orchestration tools and agent-to-agent collaboration protocols are highly technical in nature. Consequently, we need more intuitive orchestrators to realize the full potential of human-AI collaboration.

3. Leadership Implications

The integration of AI introduces a profound new dimension to contemporary leadership challenges. On one hand, it offers a significant opportunity to enhance workforce productivity; on the other hand, it has the potential to erode institutional knowledge if implemented incorrectly.

Based on my observations within organisations, the key challenges and emerging AI leadership styles focus on the following areas:

• Trust and Oversight: The natural language capabilities and confident tone of AI outputs often foster unearned trust among end users, potentially leading to confirmation bias. Consequently, it is critical that leaders establish robust institutional guardrails and formal procedures to maintain human oversight.
• Accountability and Adaptation: Cultivating a culture of adaptation requires leadership to lead by example, actively integrating AI tools into daily workflows. Simultaneously, leaders must ensure that a firm sense of accountability is maintained for all AI-generated deliverables.

Ultimately, AI provides corporate leadership teams with a generational opening to restructure and resolve organisational inefficiencies in the short term. However, successfully navigating this transition requires a careful balancing act—instilling trust across the workforce while remaining highly vigilant about imposing human oversight and strict accountability.

4. Reflection on AI Maturity

While leadership teams widely recognise Generative AI as a transformative force for business operations, a critical gap persists: many stakeholders fail to grasp that this technology necessitates a holistic transformation of people, processes, and culture—not just software stack updates. Achieving genuine organisational readiness for autonomous workflows requires comprehensive, cross-level engagement.

In my observation, meaningful change demands a top-down commitment to building trust and fostering a continuous learning mindset. I have seen this work: in my own organisation, regional leaders pioneered personal autonomous agents and demonstrated their utility during an all-hands meeting. This practical showcase was pivotal, leading to an enterprise-wide rollout. However, I recognise that even this progress represents only a partial step; establishing a fully integrated, end-to-end human-AI collaborative environment remains a significant work-in-progress.

Conversely, my research and interviews with external organisations reveal a concerning trend: many approach this transition with a top-down mandate that ignores the necessary evolution of underlying business processes. When leadership drives AI adoption purely as a cost-cutting tool without streamlining workflows, the result is often headcount reduction that fails to deliver sustainable value. This piecemeal strategy—frequently limited to basic, chatbot-style information retrieval—stalls the ROI on AI investments, causing leadership to prematurely question the value of further expenditure. The fundamental challenge, then, is not just technological deployment, but the operational and psychological alignment required to harness these systems effectively.

1. Evolution of Interaction

I started my interaction with gen AI as a searching machine. The more i was understanding the capabilities the more i was trying to instruct and train my personal AI. Now days, Claude is a kind of my thinking partner, my devil's advocate. It is the one who pushes me, challenges me and tries to fill every gap i have in my understanding of different projects and tasks. We co-create but the final judgment and accountability of any result is always on my hands.

2. Human + AI Workflow Thinking

In the future the role of AI will be closer to the one of the "right hand". Not just a tasking accomplisher but of an 24/7 colleague who is there to finish the job together with us. The human role in my belief will be and should be essential and necessary. We need to be able to validate and remain accountable for any critical outcome.

3. Leadership Implications

The upside is speed, catching small problems before they grow. But the risk I actually see, especially when I talk to clients about AI adoption, is that people start relating to AI as if it is a person, trusting it the way you should only trust a colleague. A machine cannot be held accountable, only humans can and should be.

4. Reflection on AI Maturity

No, I do not think most organizations are ready. Companies can deploy AI faster than they can decide who is accountable when it is wrong. The tech is moving fast, governance is not keeping up, and that gap is where the real risk sits.

1. Evolution of Interaction

Describe a situation where your interaction with AI evolved beyond simple information retrieval into something more iterative or collaborative. 

I spent some time with Google AI researching the effect of medicines and drugs and how medicines work over a period of time. In particular I wanted to understand how some high dosage medicines work over a period of time and bring changes in the body. 

I started with a simple prompt and giving as much information as I could. Then, I refined the prompt to ask the AI to consider reputable medical journals only. Finally, I gave more specific input about a particular situation and asked the AI to give a timeline of how the medicine works to bring changes in the body. 

All this the AI could do consulting reputable medical journals. The output was refined in a more and more iterative and repeatable manner.  

The response of the AI kept improving over multiple prompts, iterations, and repetitions. It was able to give a clear timeline of weeks over which medicine makes improvement 

What changed in the interaction dynamic? 

It was not a simple search retrieval. It was multiple iterations of prompts, questions, guiding, supplying more context and information about a particular scenario and even asking the Google Ai to refine or change the format in which the output is being given. 

The interaction was more iterative and conversational and with particular instructions and descriptions guiding the AI to come up with more detailed and refined answers that I expected. 

2. Human + AI Workflow Thinking

Identify one workflow, decision process, or professional activity where you believe AI could function as a meaningful “digital teammate.” 

Coming from the technical background, my fost familar work would be the following: 

Separate the Ai skills and agents into different specific roles that can: 

1. Data Gathering Workflow- Research the internet and gather meaningful information (e.g. for a website that shows tourist information, for popular cities, gather detailed information about particular tourist sites including amount of time to spend, what to try, any historical significance etc). We could provide the context to consider only reputable sites which rank high in google search, for example. AI is a digital teammate because the research and data gathering tasks are automated into an agent. 
2. Customer insights Workflow– On a fixed schedule, evaluate the most active and savvy leads or customers on the website, including the highest paying ones (for product and services) and based on certain categories, calculate customer interest based on activity and those that are most likely to spend more money. Then qualify them for particular promotions and send them discounts and offers. AI is a digital teammate because it qualitatively and quantitatively identifies interested leads towards paying customers. Here qualitative analysis and scoring is more important for the role of AI.
3. Aesthetic design and web design tasks – This is slightly different as this is not a scheduled workflow. AI could still be a valuable digital teammate in the initial design, aesthetic style and scheme of the website. In my experience, the initial design always needs to be refined by a human either through very specific prompts or even by making changes in many cases. 

Explain: 

• What role the AI would play 

AI would act like a digital assistant that would do all the preparatory and research work like highlighted in the 2 cases above. This cannot replace human input and human judgement because for complex or complicated scenarios, decision making would still rely with humans involved. 

• Where human judgment would remain essential 

In the above situation example, high profile, high net worth or celebrity customers would require special handling as they appreciate curated and customized attention and human interaction.  

Human judgement would still remain essential as it acts as guardrails to handle special cases, for decision making especially in scenarios which have health or money related implications, or which could lead to filing of court cases.  

For example, if providing detailed information about tourist sites which generally tends to be reputable and reliable, customers might develop an expectation that tourists given best information including precautions to take, best times to visit, places to avoid etc. There needs to be review of the information for completeness and quality. 

• How the collaboration could improve outcomes 

Collaboration could let the AI do most of the initial preparatory and data gathering work to come up with the overall content or aesthetics, for example. The human intervention could then focus on improving quality, refinement, review, and placing guardrails for ethical purposes, for example. This would lead to improvement of productivity and shortening of time, leaving the human involved to have more free time into more creative aspects of the process on how to improve the presentation, offering or other aspects. 

3. Leadership Implications

From a leadership perspective, what opportunities or concerns emerge when employees begin interacting with AI systems as collaborators rather than tools? 

For leaders, when employees and organizations start using AI, they may be concerned if the output generated from the AI has been reviewed and vetted so that accurate information is being presented and facts are checked especially high-stakes cases which could lead to court cases or monetary implications for example. Therefore, humans are accountable for verification of information presented. This is part of managing the risk of using AI. 

With AI taking up even some of the more creative tasks now, leaders may be particularly curious about variation and original human input as even creative tasks become automated (e.g. creating a music video or song) leading to a tendency to produce the same type of style of repetitive work. Variety, originality and creative human input become the differentiator in such cases to judge the quality of work 

Leaders may also worry that all employees are doing as is expected of them, i.e. using AI ethically, fact checking their work, using AI only for decision support and retaining the final authority about their work. Leaders do trust their employees and using AI across organizations increases and changes the expectations and speeds up work where maintaining guardrails and checks becomes even more important in business workflow. So, leaders may worry about managing AI risks related to confidential data, accurate outputs, and managing risks around edge cases. 

Different employees learn at different speeds there is a question if the overall skills of the employees in an organization is matched with the new AI workflows being introduced. Leaders may need to pay attention to what skills upgradation programs may that the employees are aware of how to use AI collaboratively, effectively and if every employee understands the risk that is presented if AI is not used appropriately. 

When it comes to decisions, leaders may want to place guardrails or checklists in place to ensure that enough rigor is maintained in arriving at conclusions from the facts that AI is presenting. The question is not only about decision quality but also about what processes need to be introduced to ensure reliable high-quality decisions and verification in an era where research and data gathering is fast but also the quality of sources and facts on which the decisions are based need to be checked. 

Leaders may also be focused on job redesign and what additional skills are needed by employees in a era where a lot of tasks are taken up by AI. Important questions become: 

1. What part of the workflow or business process is handled by AI?
2. What will the employees involved be incharge of?
3. How does job specification change because of the use of AI automation?
4. How is collaboration between human teammates different due to AI automating parts of the workflow?
5. Who is ultimately answerable for which part of the business workflow or decision?

4. Reflection on AI Maturity

I believe that most organizations are at different phases of starting to use AI. I can see from my company that we are only now starting to use AI as a collaborative and digital teammate that can analyze, research, and provide recommendations. Mine is a software product company and customers are other companies. 

There is a skills gap from what I see, and employees need training on what is the most effective way to use AI. Leaders are also worried about what information AI tools have access to and what is the most secure and reliable way to use these AI tools within the company without confidential or proprietary company information getting exposed. 

There is also concern about reliability of the outputs generated from AI, questions on whether the outputs be used as guidelines for improvement or whether they can be directly used as the end output after fact checking.  

From what I can see, during the initial period, where there was no comfort level with the AI tools, general preference has been to use it as a recommendation generation (to be very specific, improvement and optimization of technical code). But they have been used as recommendations with humans in charge of completely making improvements in the end products. 

So, the issue here is multifaceted – involving skills of employees, risk of data exposure, and whether the output itself from AI can be trusted and used. 

Beyond this when skill and understanding of AI matures, focus will slowly move to processes and guardrails at an organizational level to ensure accountability, work quality, and risk of data exposure. 

So, I can say that many organizations are now in the beginning or middle stages of AI integration in their workflows managing employee skills and risk of data exposure. There is still a need to institute processes to ensure reliability and quality of work, repeatability, checking and guardrails to handle the risks.  

1. Evolution of Interaction

I think the evolution started around maybe half a year ago when agentic AI got started. AI is no longer a chatbot but like my extra brain and extra hands and also could act like an agent to help me to execute a task

2. Human + AI Workflow Thinking

• One of the very useful workflow I use every day is the research agent for stock market analysis and screnning. The agent will screen the potential stock for trading Before the market opened and I need to use my own analysis and my own manual execution ( still at this moment lol).

3. Leadership Implications

Definitely I do find that my colleagues have much better output quality when using AI, at least less spelling errors and less grammatical mistakes. However one of the concern that I saw from their work over the past few months is that sometimes they may not even know the complicated logica and answer from the A.I I'm afraud that gradually they may lose the ability to have their own judgment

4. Reflection on AI Maturity

Yes or no and it very much depends on the vision from leadership team. I think you like or not, the company that could adopt AI effectively and efficiently , they will have a higher chance to win the game.

Describe a situation where your interaction with AI evolved beyond simple information retrieval into something more iterative or collaborative.

As a career coach, one of the key services that I deliver for my clients, is advising them on the competency match between their resumes and the job description. With AI, that matching can be done instantaneous and it could offer to help you update your resume to be a fit with the job description. The problem is the authenticity of the of the content and the subsequent mismatch between resume and physical interview. 

Identify one workflow, decision process, or professional activity where you believe AI could function as a meaningful “digital teammate.

One aspect is the pre-intake of clients before every session. For career coaching, we do ask clients to answer questionnaires to have a better sense of the clients' interests and strengths. AI as digital teammate could be used to automate this process and provide a preliminary assessment of the responses to help career coaches perform the first cut of the intake. 

From a leadership perspective, what opportunities or concerns emerge when employees begin interacting with AI systems as collaborators rather than tools?

As a HR leader, the concerns i have revolves around the loss of human touch/empathy vs efficiency brought about by AI. Not long ago, organisations believe that people drive culture and culture drives high performance in an organisation. With AI being a collaborator and increasingly, "humanified as seen in suno and claude", it makes me wonder whether people will get out of touch with each other as many of the processes including frontline interactions are powered by AI. 

From an opportunity standpoint, AI will bring about efficiencies and greater capacity for humans to discard mundane tasks and focus on impactful work and with the help of AI, do these work better. 

Do you believe most organisations today are psychologically and operationally ready for collaborative AI systems?

Why or why not?

Interestingly, i just had a discussion with my teams on how to embark on AI and sure that we are aligned across the entire HR function. One of the starting points that i made was that yes, there is definitely hype on AI and people including leaders, see AI as the next bound of transformation which is imperative to hope on the AI train in order not to be left behind.  I dont dispute this point as AI does increase efficiencies and open doors to unlimited potentials which previously we dont think we can do it as non-IT trained persons. On the other hand, to operationalise AI, we need to have a holistic plan - are our processes AI possible (do we need to tweak our processes in order for AI to be optimise), what are the data that are found in these processes - are they sensitive, what are the cost and ROI on AI investment, Is it sustainable in the long run and not forgetting, the environmental aspect of AI, stakeholder experience and impact (not just to the clients but to the processes of other teams) and the human readiness (are they trained so that they are not afraid of AI)? i ended up started by getting my team AI certified, review the SOPs and then later on, followed with a matrix table with all the above points, as a start, to decide whether to AI a certain processes or not. 

1. Describe a situation where your interaction with AI evolved beyond simple information retrieval into something more iterative or collaborative. What changed in the interaction dynamic?

One situation where my interaction with AI became truly collaborative was when I used an AI agent to support demand forecasting.

Instead of simply asking AI to generate a forecast, I worked with the AI agent iteratively. At the beginning, I carefully reviewed the outputs to ensure the data was accurate and consistent with historical trends. Whenever I found discrepancies, I provided feedback, corrected the assumptions, and asked the AI to explain its reasoning.

Through this iterative process, the AI gradually produced more reliable forecasts because I continuously refined the prompts, clarified the business context, and validated the results. Rather than acting as a tool that simply returned answers, the AI became a collaborative partner that helped analyse data, generate forecasts, and improve its outputs based on my feedback.

What changed in the interaction dynamic was my role. Initially, I treated AI as a tool for generating outputs. Over time, I shifted to acting as a reviewer and coach—providing context, validating results, and guiding the AI toward better performance. At the same time, the AI evolved from answering individual requests to becoming an active collaborator in the forecasting workflow.

2. Identify one workflow, decision process, or professional activity where you believe AI could function as a meaningful “digital teammate.”

One professional activity where I believe AI could function as a meaningful digital teammate is creating business presentations.

In my role, I frequently prepare presentation decks for internal stakeholders and business partners. I usually start by briefing ChatGPT on the objective of the presentation, the target audience, the key messages, and any supporting data.

Based on this context, ChatGPT suggests a logical deck structure, key content for each slide, and ideas for the visual layout. I then review the proposed outline, provide feedback, refine the messaging, and ask for alternative approaches until the presentation aligns with my objectives.

What makes this collaboration valuable is the iterative feedback loop. Rather than simply generating slides, AI helps me organise my thinking, challenge my ideas, and explore different ways to present the story. Meanwhile, I contribute business context, validate the recommendations, and make the final decisions on the content and design.

In this workflow, AI acts as a creative partner that accelerates the presentation development process, while I remain responsible for ensuring the deck is accurate, relevant, and persuasive for the intended audience.

3. From a leadership perspective, what opportunities or concerns emerge when employees begin interacting with AI systems as collaborators rather than tools?

From a leadership perspective, AI collaboration creates significant opportunities but also introduces important responsibilities.

One of the biggest opportunities is increased productivity. AI can help employees complete repetitive and time-consuming tasks more quickly, such as preparing reports, analysing data, or creating presentation drafts. This allows employees to spend more time on higher-value activities such as strategic thinking, problem-solving, relationship building, and innovation.

However, there are also important concerns. Employees should not rely on AI to generate ideas or make decisions independently. Instead, they need to provide clear direction, business context, and objectives before working with AI. Once AI produces an output, employees must critically evaluate and validate the results, as AI can still make mistakes or generate inaccurate information.

From a leadership perspective, success depends on creating a culture where AI augments human capabilities rather than replaces human judgment. Leaders should invest in AI literacy, establish clear governance, and ensure employees understand that accountability for business decisions always remains with people, not AI.

4. Do you believe most organisations today are psychologically and operationally ready for collaborative AI systems? Why or why not?

I believe most organisations have recognised the importance of AI and have started their AI transformation journey. However, many are still facing several bottlenecks that limit the successful adoption of collaborative AI systems.

• Uneven AI literacy and skills

  Employees have different levels of AI knowledge and proficiency. While some are comfortable using AI tools, many others still lack the skills to effectively integrate AI into their daily work.

• Undocumented and inefficient business processes

  Many organisations have established business processes, but these processes are often poorly documented. Without a clear understanding of the workflow and its bottlenecks, it is difficult to identify where AI can create the most value.

• Unrealistic expectations of AI

  Many people believe AI can do everything—from writing code and preparing reports to designing images. In reality, AI still has limitations and can generate hallucinations or inaccurate outputs. For high-impact decisions, human-in-the-loop oversight remains essential to validate AI recommendations and make the final decisions.

• Evolving AI regulations and data privacy concerns

  AI regulations are still developing in many countries. This creates uncertainty around governance, compliance, and the risk of data leakage when organisations adopt AI solutions.

• Rising AI costs

  The cost of using AI is increasing as organisations scale adoption. Without proper governance, user training, and cost management, AI expenses can eventually exceed the cost of having humans perform the same tasks.

My interaction with AI tools has become more collaborative with increased usage. This is especially evident in presentations to senior executives where I would use AI tools to validate my key messages, story / presentation flow, slide content and graphics. The back and forth iterations using AI feedback has made my presentations improved significantly while saving time in the review process. This highlighted a much more collaborative interaction dynamic with AI. AI had become my digital team mate bouncing ideas and iterating on the presentation. 

I do admit that although AI was a big help, personal critical thinking and human judgement in navigating how the story and key message will land with the audience is still a key deciding factor in making the outcome a success. My judgement in the tone, the gestures and human touches are still essential. 

Ultimately, I do see that AI as a critical tool in enabling productivity and improving decision quality. However, I do feel that it is still tool. Though it can be a collaborator, the ultimate outcome / accountability owner still has to be human. 

I believe most organisations and people are not ready for the impact of AI systems. Right now, I believe organisations are simply creating new AI test cases and showcasing benefits. The true benefit is to rip out the existing structures and redesign with AI. In that world, I believe I may also not be entirely ready for.  

Fundamentally, I am concerned that there will be far fewer jobs. I do acknowledge that jobs will evolve and new kinds will be created due to AI. However, many white collar functions that we know today may no longer be needed. Optimistically, I believe this may take a few years when leaders redesign processes and organisations with AI as its core. 

Evolution of my interaction with AI : My first interactions with AI was its infancy when we used mathematical models to predict the next few moves of the markets. At that point in time, computing power was expensive and hence such tools were in the hands of very few corporations. Now hyperscalers are in the process of democratising the usage of AI tools. I am in awe of the developments of technology in this area and it’s upto me to build a better collaboration with this technology. Clearly there is a long road ahead for me in this.

Human + AI workflow thinking: many financial firms have a repository of the regulations that are applicable to its business. As a digital teammate, I would like AI to not just react to my query for specific regulations but also highlight areas that may be related but i have not asked for specifically in my query. Effectively, it could help design financial products where human judgement remains essential and the collaboration is powerful to deliver a better outcome in terms of solutions to cater to specific contexts. This, to my mind, helps individual contributors. From a team managers perspective, i feel introducing Agentic AI for further workflow/process automations comes with a human cost of job redundancy and so i have yet to apply mind deeply into this area.

Leadership implications: 

1/ need to encourage AI adoption - training is essential

2/ cost implications of tokens

3/ information security aspects - do not want proprietary information to be available to the AI companies

4/ do we develop our own LLM and SLMs in-house?

5/ ensuring consistency of output that is auditable.

6/ impact on white-collar as also creative jobs. High possibility of job redundancies.

AI maturity : 

I work in a US bank in Asia geography. I see high AI maturity in my organization, but i do not see the same in Asian banks. It is similar to the use of using financial derivatives for hedging risk and market making in them. I guess that this is natural with early adopters and laggards of technology.

1.Evolution of interaction

Initially, I used AI primarily as a search engine and a tool for information retrieval, content extraction, and document refinement. The interaction was straightforward: I asked questions, received answers, and used the information to complete my work.

The dynamic changed when I started using AI for creative tasks. A memorable example was creating a reunion song for my 1990 high school classmates using SUNO AI. Instead of simply providing information, the AI became a creative collaborator. I shared the theme, emotions, and memories I wanted to capture, reviewed the generated lyrics and music, provided feedback, and refined the output through multiple iterations until it matched my vision.

Beyond creative tasks, I also experienced AI's collaborative value in my professional work. During the evaluation of ERP solutions for our manufacturing operations, AI significantly accelerated the process of comparing different options. By consolidating inputs from both functional and technical teams, AI helped organize, compare, and quantify the findings in a structured manner. It enabled us to identify strengths, gaps, and trade-offs much more efficiently than traditional manual analysis.

What would normally have required extensive effort to collect information from multiple stakeholders, consolidate it into spreadsheets, perform comparisons, and prepare evaluation reports was completed in a fraction of the time. Instead of spending most of the effort on compiling data, the team could focus on validating assumptions, refining the evaluation criteria, and making informed decisions.

This changed my interaction with AI from simply asking questions to working with it as an analytical and collaborative partner. AI became an active participant in the process , helping synthesize information, structure complex comparisons, and support better decision-making through an iterative cycle of feedback and refinement.

2.Human + AI work flow thinking

A good example is the Accounts Payable process, where AI can act as an intelligent assistant rather than simply automating tasks. The AI can compare supplier invoices against the corresponding purchase orders, packing lists, and material receipt records to verify that quantities, prices, and other key details are consistent. It can then highlight discrepancies, identify exceptions, and provide insights into potential issues that require attention.

Instead of manually reviewing multiple documents line by line, the Accounts Payable executive receives a summarized analysis of the findings. They can then review the identified exceptions, validate the AI's recommendations, and make the final decision before approving payment to the supplier. This transforms the process from a time-consuming manual verification exercise into a collaborative workflow where AI performs the initial analysis and the human provides judgment and final approval, improving both efficiency and accuracy.

3.Leadership implications

Successful AI adoption delivers value beyond automation. It drives fundamental improvements in the way organizations operate by:

Process Standardization: Encourages the adoption of standardized business processes across departments, reducing variations and improving consistency.

Process Confidence and Reliability: Establishes reliable, repeatable, and data-driven processes, increasing confidence in business operations and decision-making.

A New Approach to Process Design: Shifts the focus from designing processes for manual execution to designing them for automation, intelligence, scalability, and continuous optimization.

Better Resource Utilization: Frees employees from repetitive and routine tasks, allowing them to focus on higher-value activities such as problem-solving, innovation, customer engagement, and strategic decision-making.

Elimination of Non-Value-Added Activities: Identifies and removes unnecessary manual work, duplicate data entry, and inefficient approval cycles, resulting in faster processes, lower operational costs, and improved productivity.

4.Reflection on AI maturity

Many organizations are eager to adopt AI, but they often underestimate an important reality—they are not yet AI-ready.

AI adoption does not begin with a budget, a technology platform, or a desire to use the latest innovation. It starts with a clear understanding of the business problem, the organization's current maturity, and the prerequisites required for AI to deliver meaningful value. AI is not a solution that can compensate for inconsistent processes, poor-quality data, or a lack of standardization.

For example, consider a make-to-order manufacturing company serving around 30 customers, each with multiple SKUs. The costing process is performed by different teams, each following its own methods, assumptions, and exception-handling practices. In such an environment, introducing AI is unlikely to produce reliable or consistent results because the underlying process itself is fragmented.

The first step should be process optimization and standardization. The organization must define a common costing methodology, establish standard workflows, and minimize unnecessary variations while clearly documenting legitimate business exceptions. Only after these foundations are in place can AI effectively automate, analyze, or optimize the process.

Similarly, many organizations are still in the early stages of digital transformation, moving from spreadsheets and manual processes to an ERP system. Before investing in AI, they should complete their ERP implementation, stabilize business processes, and establish a single source of truth for enterprise data. Without accurate, consistent, and well-governed data, AI outputs will be unreliable and may lead to poor business decisions.

Successful AI adoption is therefore not just a technology initiative—it is a business transformation journey. Organizations that first standardize their processes, improve data quality, and strengthen their digital foundation are far more likely to realize the full benefits of AI than those that adopt it prematurely.

1. Evolution of Interaction

At first, I didn't like when Google was providing the AI mode when I had asked a simple Google search request. I found it unnecessary and I was worried about the cost of it. Then I realised for some of my searches that the AI mode delivered a well-structured response, with diverse point of views and links to references, which was how I learnt to methodically or scientifically explore a topic. So that's when I started to rely more on Gemini when I had a topic I wanted to learn or clarify about. I even got a free 1 yr license to Perplexity pro, so I was able to explore the newest feature (at the time) of reasoning. I really enjoy reading how the AI was reasoning and fundamentally it helped me trust it more and wanting to use it more.

2. Human + AI Workflow Thinking

I'm the one usually planning for family activities and holidays. I started using AI as a workflow that ultimately supported my decision to a destination, itinerary and activities. The AI is most efficient in narrowing down the destination based on time of the year, weather, budget and the type of holidays we want as a family. Once the destination is identified, I would usually use the AI for refining the type of activities and how a day is organised overall, with a mix of restful moments and activities. I get usually everything down to 2-3 recommendations and would make the final call. Where I have kept my human judgement is whenever booking/payment is required. I always verify accuracy as sometimes a place is temporarily or permanently closed and the AI hasn't picked it up. Overall, I can plan for my holidays within a day, while before it could take me weeks of research.

3. Leadership Implications

More and more employees from my company are using AI to be more productive for themselves. As our industry is highly regulated, from day one our AI engine was private and it was forbidden for us to use anything external to our company's Tech stack. It helped trusting the AI to handle sensitive and confidential information, supporting staff to explore further use of AI in their day to day job.

Because the company is big, it took the effort of a digital transformation strategy to re-design our business processes and embed AI by design rather than retrofit it. This is definitely the right approach, but it takes time, and while this is being delivered, employees continue to use it without clear guidelines, creating their own workflows, connecting siloed dataset without embedding standards etc. This leads to lengthier workforce adaptation when the strategic approach comes in and a more comprehensive change management, costing more to the company. I don't think there's a "right" approach here, as it is also necessary to encourage employees to explore and innovate with AI. However, as leaders, we should be more mindful of clearly stating transition plans with clear and transparent milestones.

4. Reflection on AI Maturity

I believe there's a mix within organisations but also across countries. Some governments, like in Singapore, have set the tone with clear AI objectives and AI framework, supporting companies to deliver on their AI objectives with incentives and by ensuring workforce readiness through cheap/free AI courses. Whereas some are lagging behind with lack of infrastructure and governmental support.

Generally, companies have started using tools like Copilot, which is an easy entry point to get used to having a digital teammate. They are now looking into embedding agentic AI, which requires stronger framework and accountabilities from leadership teams. These leadership teams need to be trained first, but I'm not seeing that happening as a priority. It is mostly delegated to IT department or digital and analytics teams, which I believe is not sufficient for a successful adaptation at scale within a same company.

I have been experimenting with AI for the past few months.

Initially, my use of AI was mainly focused on content generation. For example, I would provide a few lines describing the purpose of an email and ask AI to draft it. The interaction was simple and transactional. I provided the context, and AI generated the output.

Over time, this interaction has become more iterative and collaborative.

One example is a biweekly report that I prepare. It brings together inputs from different sources and requires interpretation, correlation, and identification of useful insights. I explained to AI how I normally look at the data, what patterns I consider, and how I derive conclusions.

I then converted this approach into a set of instructions. AI now helps turn raw data into a more structured and insightful report. This has reduced the effort required and improved consistency. However, I still review the output, validate the conclusions, and add the context that AI may not understand.

A workflow where AI could act as a meaningful digital teammate is business analysis and reporting.

AI could help organise information, identify patterns, highlight gaps, and suggest areas that need attention. Human judgment would still remain essential in interpreting the findings, understanding the wider context, and deciding what action to take.

The collaboration could improve outcomes by reducing the time spent collecting and structuring information. It would allow people to focus more on judgment, decision-making, and action.

From a leadership perspective, I also see some concerns.

Many employees are using AI to generate content, but the output often becomes scripted and generic. In one case, I asked my team to write a review, and much of what I received appeared to be standard AI-generated content. It was not necessarily wrong, but it lacked individual thinking, specific observations, and personal judgment.

This creates a risk that the output looks polished, but the thinking behind it is limited.

Trust and accountability therefore become important. Leaders need confidence that the AI output has been reviewed and validated by the individual using it. Accountability must remain with the person making the decision.

This is especially relevant in industries where safety, reliability, and quality matter. If AI contributes to an analysis or recommendation, the key question is whether the output has been reviewed by a qualified person and whether the underlying data is reliable.

AI may also rely on outdated or incomplete information. It may present an answer with confidence even when the evidence is weak. This makes human oversight essential.

From an organisational perspective, I do not believe most organisations are fully ready for collaborative AI.

There is still uncertainty around trust, data privacy, intellectual property, and accountability. Many organisations do not yet have enough clarity on what information employees should enter into AI systems, how that information will be used, and how outputs should be validated.

Until these issues are clearer, adoption will remain uneven. Some employees will avoid AI because they do not trust it, while others may use it without enough caution.

The real value will come from combining the speed and pattern recognition of AI with human judgment, context, and accountability.

1. Evolution of Interaction

Working in the construction industry, I often deal with technical terms, contracts and regulations. Initially, I used AI mainly to explain unfamiliar concepts and summarise information.

Over time, my interaction became much more collaborative. Instead of only asking questions, I began using AI to improve emails, compare documents, prepare presentations, and brainstorm different approaches to problems. The interaction shifted from simple information retrieval to an iterative discussion where I refine prompts and AI helps me explore different perspectives before I make the final decision.

2. Human + AI Workflow Thinking

I believe contract administration and commercial correspondence are ideal areas for AI to act as a digital teammate. AI can review documents, identify relevant clauses, draft responses, and summarise meetings.

However, human judgment remains essential when interpreting contractual intent, managing stakeholder relationships, negotiating, and making commercial decisions. AI improves efficiency, but people remain responsible for the final outcome.

3. Leadership Implications

From a leadership perspective, AI presents opportunities to improve productivity, decision quality, and knowledge sharing. At the same time, organisations must ensure employees do not rely on AI blindly. Leaders need to establish clear governance, encourage critical thinking, and help employees develop AI literacy so that AI becomes an enabler rather than a replacement for human expertise.

4. Reflection on AI Maturity

I think most organisations are interested in adopting AI but are not yet fully ready for collaborative AI systems. Many still face challenges with data quality, governance, and employee readiness.

In my industry, AI is already valuable for drafting documents, reviewing contracts, and preparing reports. However, decisions involving safety, compliance, and commercial risks still require experienced professionals. The organisations that will benefit most are those that treat AI as a trusted digital teammate while keeping humans accountable for final decisions.

1. Evolution of Interaction

I started using AI only a year ago. Generally, for engineering technical review and documents. I am a professional civil engineer, and my work is more on engineering construction management. I would say my usage of AI is more as I find using it useful in my work. 

2. Human + AI Workflow Thinking

I strongly believe in my professional career, using AI is very useful and will provide a faster workflow. Having myself as the user and AI thinking, iterating and reviewing together is a good combination. But of course, I need to be specific in asking the AI and requesting for more related information that AI can offer to me.

3. Leadership Implication

In my work, AI will provide more work for us engineers, but again the leaders of my management need to appreciate the AI to invest and to implement to our projects. That is yet to be seen. I don't think engineering and construction companies have heavily invested to AIs in Singapore.

4. Reflection on AI Maturity

AI is the generation. AI is the future. AI must be used. Its maturity is rapid. 

1. Evolution of interaction

In my line of business which is Travel, I need to create ads for posting on FB. I would use Gemini and they create it for me, and I request gemini to provide the photos. If I am not happy, I would ask it to redo until I am satisfied.

2. Human  + AI workflow thinking

AI and my team would work together on developing tour packages in countries we are not familiar with. It would be a collaboration with my team and AI as we do not exactly follow all of its suggestions. We have to tailor fit it to the interests of the Filipino people. Our culture must be taken into consideration in developing tours. AI would only give us ideas and suggestions but the final tour to be sold will still be my teams decision.

3.Leadership implications 

With AI , I trust my team to be truthful with their inputs, anyway I will still have the final word on our tour packages , inclusions and layout. The team is made accountable on their submissions, as I would always remind them to be creative and think beyond the box. The final decision is made by me with the recommendations of the team. Unlike before, tour packages take quite a long time as they rely on our foreign  tour operators . Team dynamics have changed as they now work more closely unlike before AI came into our operations, The team i would say is very much ready in adapting AI as we have been using it for almost a year.

4. In my industry, I would say that Filipinos are not psychologically  and operationally ready for AI integration in their companies. There is still resentment that AI will take out their importance in the company and that with simplified AI systems, they will lose their job.

An example here in the Philippines  is the focus of Industry leaders and public officials who historically advocated the the Filipino tourism brand is about love, warmth and dedication which AI cannot simulate.