HR Tech Europe Meetup – Munich Edition
Leveraging AI & Automation Responsibly for Impact & Experience
with Mercer and Elke Manjet

This document aims to provide a summary of our conversation and inspire our further journey with AI.

1 - Trust & Transparency

• Transparency has always been a sign of a good company culture – and should be applied to people decisions. Trust in the decision process increases trust in the company, even more as AI technology increases uncertainty for many people.
• The absolute key question for companies to answer along with AI implementation is: where do we want to have the human in the loop? People decisions should always be taken with human oversight (e.g. hiring, promotions…) even if the technology can take them autonomously.
• Expectation management is key – things don´t change over night if you plug in AI. Per examples shared implementation did not go well, if no sufficient focus was given to a strong communication, enablement and adoption plan.

2 - Operational Impact of Agentic AI

• We are seeing highest adoption of AI in Service Delivery. Talent Acquisition is the 2nd highest area with AI adoption in HR.  As this point these are the areas where highest efficiency gains are realized. Services teams will move from providing basic services to shaping employee experience and TA will focus on creating great experiences for candidates.
• Companies are already establishing agents that take on HRBP tasks (e.g. IBM) like e.g. career consultation, preparing managers for yearly compensation review. They admit that key KPIs like employee satisfaction have dropped at the start of the transformation, however, can absolutely be elevated as AI usage matures and adoption expands. 
• Along with the business need to have the right skills for strategy execution, skill platforms providing transparency into skill gaps, suggestions to close them, and guiding next growth steps, are being implemented more. This will have a significant impact on the teams owning learning and development.

3 - Vision for HR in an agentic world

• What we know today is that jobs in HR will significantly change. We can look at this change as an opportunity and shape the future – accordingly we should not consider AI a further project but leverage the opportunity to fundamentally rethink how we work.
• Freed from many operational tasks HR can finally be the business partner, impacting business outcomes (e.g. productivity) with a holistic and dynamic workforce plan combining build (skilling), buy (hiring) and borrow (contingent workforce) strategies to deploy the right skills where they are most needed.
• HR should focus on (re-) building culture, aligned with the company´s future, reflected in key decisions and everyday interaction, as well as in the key people processes.
• HR must own the redesign / reinvention of work in the HR function and the company. Decomposing jobs to tasks, deciding what continues to be done by humans and what by agents, is a big undertaken. This will require redesign of how we hire, how we define careers, how we continuously skill and deploy people and how we create employee experiences. It will additionally require strong focus on transformation to manage the implications on people and leave nobody behind.

4 - How should HR leaders redesign work when AI systems act as “colleagues,” not just tools?

• Agentic AI’s dual nature as both a tool and coworker poses challenges, that current management frameworks are not prepared to deal with. When a tool malfunctions, it’s a defect. When a coworker makes a mistake, it’s a management and learning opportunity. We need to apply more of a coworker mindset to AI – agents need to be trained and supervised. Most important - there must be a governance model in place to regularly check outcomes – that ´judgement call´ remains on us humans. The risk that AI produces wrong answers or derives wrong conclusions is always there.
• A big question in this context is how are we redeploying freed time? How can we best use the time we gained to create more value for the business and more impact as a function. The answer obviously is very dependent upon where an HR team is in its maturity and what the business context requires.

5 - What does “AI literacy” look like for HR - not just conceptually, but practically day-to-day?

• First it is important from a cultural perspective to embrace a beginners mindset – with AI we are all on the journey and we´re all learning as we go – it takes experimentation, sharing and storytelling.
• For leaders it is important to admit they don’t have all the answers.
• The key skills to interact with AI are prompting, iterative thinking and judgement. Typically, you start interacting with AI by prompting, then the next step is to iterate to get to a better outcome, the third skill is judgement – assessing how reliable the outcome is, how relevant and how current.
• What gains more importance in the AI age are the `human` competencies e.g. emotional intelligence, creativity, judgement – they are crucial to include in our skilling focus, as relevant as technical skills.

6 - How do we ensure AI reduces bias rather than amplifies it?

• By regularly reviewing the outcomes make sense and are not biased. Given LLMs are trained with data freely available in the web, these can be outdated and can amplify past bias. If the AI is trained with your company internal data, it really depends on the quality of your data as well as a clear reflection and acknowledgement of what needs to change.
• As participants addressed from their own experience, working on harmonized, high-quality data is the foundation for a successful AI journey, and often a significant project of its own.

7 - What is the most important decision organizations must make now to avoid falling behind in 12–18 months?

• If HR teams have not implemented AI yet they are already behind and should quickly build their strategy on what pain points they want to solve resp. what strategic advantage they want to gain with AI.
• Starting as early as possible, setting clear goal, learning fast and being prepared for some ´valley of tears´ before the benefits become visible / tangible in the company is an experience frequently shared from AI projects.

Written by Elke Manjet