More AI tokens, or more thinking and learning?

Karen Clark shares her thoughts.

Last week I was challenged, constructively, to explain why an investment in PARC Access is worth it when AI offers a route to “so much insight and knowledge”.

It isn’t the first time I’ve been asked, and I suspect it’s a question on the minds of many reward leaders as they set budgets for next year. Why invest in PARC, or any specialist partner, when you’re under pressure to save costs, cut budgets and embrace the efficiencies of AI? Function-specific spend is often the most exposed. Larger items tend to be tied into contracts longer than twelve months, and organisation-wide investments are usually easier to justify to colleagues in finance and procurement.

It’s a fair challenge, and one I’ve taken time to reflect on. If your team needs faster summarising, modelling, drafting or first-pass analysis, AI will, depending on how it’s accounted for, most likely win on cost and speed.

But in my time at PARC, and in conversations with practitioners across our community, speed of answer or slide deck production has rarely been what holds a reward function back. (I know AI does far more than produce slide decks, but allow me the shorthand!) Pace matters, of course. But reward is valued for its expertise, its advice and counsel, and the quality of its contribution to the matters that concern Boards and other stakeholders.

And the questions reward is being asked to answer have changed. Fairness, transparency, purpose and legitimacy now sit alongside the technical work, and RemCos, investors, regulators, unions and employees are all watching closely. AI can give you a fluent, well-organised answer to a well-constructed question. What it won’t do, unless prompted by a highly discerning and curious user, is ask whether you’re asking the right question in the first place.

I think that matters. AI works within the framing you give it. Ask it to help design a better bonus scheme and it will help you design a better bonus scheme. It is far less likely to ask whether a bonus is the right answer at all, or whether the assumption that more money drives better performance actually holds in your organisation. Reward is full of assumptions that have been inherited, rarely examined and built into how we design pay. Some still hold and some don’t. Being able to tell the difference is where the real value lies, and it’s very hard to do on your own, or with a tool that tends to agree with you.

This is where I believe PARC earns its place as an external partner to reward teams and their HR colleagues. Not by handing you ready-made answers, but by giving you a forum where your thinking gets tested. Senior peers who will push back. Academics who bring behavioural science, economics and ethics into the conversation. The chance to hear what comparable organisations are actually working through, and to rehearse difficult conversations before you have them for real, in your own context and with your own stakeholders. I know some organisations are getting real value from rehearsing with synthetic personas, but I’m talking about conversing with real people!

It’s also important to us that learning and curiosity sit at the heart of what we do. I’ve written before that learning in reward is too often treated as a perk, the first thing to go when budgets tighten. That’s a false economy, and AI makes it more so, not less. AI makes a capable team faster. It can also make a less experienced team confidently wrong. I’m not the first to use that phrase, but it resonates with me. I can think of moments in my own career, long before AI, when I was confidently wrong and paid the price for it. The protection against that is judgement, and judgement is built through learning, hard graft and the occasional failure. There is real value in working through live problems, being challenged, getting things wrong in a safe space, and learning alongside people facing similar challenges in different contexts. That kind of peer learning is very hard to create internally, and enormously valuable when you find it.

At our conference this year, we asked why more investment in technology hasn’t delivered more productivity. A reward team with more AI, operating with the same thinking, is a small version of that paradox.

So for me, the question isn’t PARC or AI, or indeed any trusted partner or AI. Use AI to do carefully chosen work faster. Use trusted partners to make sure it’s the right work and to challenge the thinking behind it. Use them to build a team that can create, deliver and stand behind its recommendations, to sharpen critical thinking, communication and influencing skills, and to get ahead of the issues that don’t come with an AI playbook, or that matter too much for anything less than a solution built for your organisation.

And don’t underestimate the cost of using AI as a substitute for tailored, professional support across insights and learning. The value of a good external partner compounds over time, as your team’s thinking and judgement grow, and that capability is what will serve you best when the next round of cost pressures arrives.