We Don't Have an AI Problem. We Have a Forgetting Problem.

 

What happens when the real bottleneck in consulting isn't judgement; it's follow-through

By Rasiah Keerthana | MBA Batch 37 | Jul 2026


“Still done by hand — for now.” Source: [Rasiah Keerthana] (2026). Personal photograph. Asia Pacific Group, Kandy, Sri Lanka.


Two things are true at Asia Pacific Group right now.

One: everything is still manual; we check every profile ourselves; nothing is automated. Two: the biggest recurring problem isn't a bad decision; it's staff simply forgetting. A task gets assigned, marked urgent with a clear deadline, and still gets missed. We call to follow up, and the answer is simply, “I forgot.”

That is a workflow-memory problem, exactly what agentic AI is built to remove. Generative AI drafts something when asked; agentic AI plans and executes a workflow autonomously, without a human needing to remember it (Aisera, 2026). Sri Lankan HR functions are only beginning to apply this structurally (Business News LK, 2026), and firms still doing everything by hand are the clearest case for where it would help first: not shortlisting, just follow-through.


“The shortlist used to take two days. Increasingly, it doesn't need a human to build it at all.” Source: [Jakub Zerdzicki], Pexels (2026).


But automating the forgetting away raises an uncomfortable question a talent audit (Henderson, 2017) forces you to ask: are we solving a capability gap, or hiding one? A talent audit means honestly assessing skills and behaviors against what the role needs. A task missed despite being called and flagged urgent may not be a memory failure; it may be a sign the task was never truly owned. Letting AI catch every dropped deadline removes the pain, but also the evidence HR needs to see who is not taking ownership. A fair counter-argument: if an agent chases every deadline, staff may stop treating ownership as their job at all.


“What AI can't audit: whether someone's ready for more.” Source: [Rasiah Keerthana] (2026). Personal photograph. Asia Pacific Group, Colombo, Sri Lanka.


Most of the team never has this problem; the gap shows up in a minority of cases, which is why it is worth auditing rather than assuming it is everyone. When a gap in ownership starts affecting client service, we recruit; client trust is not something we gamble on while one person's gap gets sorted out. The redesign question is not only what AI can do for us. It is what the audit reveals once AI closes the memory gap, who is ready for judgement-based work, and who needs a new hire instead. Automating the follow-up is easy. Deciding who stays, who grows, and who gets replaced is the harder decision.

So here's my question to you: if AI took over every reminder your team currently needs, would you find out who's ready for more responsibility or just lose the last piece of evidence that someone wasn't?


Watch: How AI Is Revolutionizing Recruitment in 2026 and Beyond

Watch: Agentic AI Explained - McKinsey & Company


References

Aisera (2026) AI recruitment: the 2026 guide to agentic AI and hiring. Available at: https://aisera.com/blog/ai-recruiting/ (Accessed: 12 July 2026).

Business News LK (2026) The future of work: how AI is transforming human resource management in Sri Lanka. Available at: https://businessnews.lk/2026/07/03/the-future-of-work-how-ai-is-transforming-human-resource-management-in-sri-lanka/ (Accessed: 12 Jul 2026).

Henderson, I. (2017) Human Resource Management for MBA Students (3rd ed.). London: CIPD.

HR.com (2026) From reactive to predictive: how agentic AI is rewiring payroll, HR, and the future of work. [Video]. YouTube. Available at: https://www.youtube.com/watch?v=3o9ENCkO1ow (Accessed: 14 July 2026).

McKinsey & Company (2025) Agentic AI explained. [Video]. YouTube. Available at: https://www.youtube.com/watch?v=DsRNp3cmJm0 (Accessed: 14 July 2026).

MeetGeek (2026) How AI is revolutionizing recruitment in 2026 and beyond. [Video]. YouTube. Available at: https://www.youtube.com/watch?v=eAr2kpPNUcI (Accessed: 17 July 2026).

Nawaz, N., Arunachalam, H., Pathi, B.K. and Gajenderan, V. (2024) 'The adoption of artificial intelligence in human resources management practices', International Journal of Information Management Data Insights, 4(1), 100208.

TheHireHub.AI (2026) AI superagents in HR: the 2026 tipping point for recruitment. Available at: https://www.thehirehub.ai/blog/ai-superagents-hr-2026-tipping-point (Accessed: 17 July 2026).




Comments

  1. A thought provoking perspective on AI adoption in HR. I agree that AI can reduce administrative gaps and improve follow-through, but it should complement rather than replace human judgement. The challenge is ensuring technology helps identify capability gaps instead of hiding them. A strong balance between automation, accountability, and employee development will be essential.

    ReplyDelete
    Replies
    1. Thank you; that's exactly the tension I was trying to name. I did add that the risk isn't only tech hiding gaps; it's that automation can look like a fix even when the underlying gap is still there, just less visible day-to-day. So the balance you mention probably needs a deliberate check built in, not just good intentions about using AI responsibly.

      Delete
  2. I agree that technology can improve efficiency. But, organisations should also understand the underlying capability gaps before automating processes.

    ReplyDelete
    Replies
    1. Thank you. I agree, and the order really matters here. Automating before understanding the gap risks just making a flawed process faster, not better. In practice, though, organizations rarely pause to diagnose first, so the real challenge is building that diagnostic step into the decision, not just agreeing it should happen.

      Delete

Post a Comment

Popular posts from this blog

"We Help the World Move — But Can't Keep Our Own People"

Your Best Employee Just Got Bored. Now What?

AI Just Shortlisted Your Next Employee. Did Anyone Ask HR?