|
Can you believe there’s a full-blown festival in Berlin dedicated entirely to HR? It’s called EMBRACE Festival and is equal parts business conference and DJ party. How very Berlin-esque: Where else would you find 1,000+ leaders wrapped in street-food trucks wrapped and techno vibes?
I had the honor of being the opening speaker for Embrace 2025, speaking on how technology is reshaping HR rapidly but also how this pace affects the human brain on a biological level, and what that means for recruiting, mental health, and leadership.
One thought I shared there feels especially important to pass on here:
Ask yourself, would you ever hire a new employee without giving them a job description, an onboarding plan, or regular feedback?
Of course not. And yet, that’s exactly what most companies are doing with their AI agents.
We’ve made AI incredibly powerful, but now comes the real challenge: teaching it what we want.
The secret? Start treating AI like a new team member.
In all seriousness, imagine AI systems like a human intern that needs your help in ramping up.
Here are 6 lessons to remember when “onboarding” AI in your organization:
 Structure matters: Give every AI agent a clear onboarding process. define its role upfront, e.g. “draft first versions of client proposals as if you were a professional with 20+ years of experience, but keep it streamlined and to the point”
Details count: Share data, workflows, and context, like you would with a colleague. Upload templates, FAQs, and past documents so the AI learns your language and standards right away.
KPIs are key: Set expectations and measure outcomes. Decide in advance whether success means faster turnaround, higher quality drafts, or reduced error rates. And be fair with the AI that it will only ever be able to optimize for your KPIs as good as your instructions and boundaries are. You have to choose this.
Feedback loops: Build mechanisms to improve performance over time. Regularly review outputs, highlight what works, and retrain with corrected examples just like coaching a new hire.
Leadership gap: Most AI failures aren’t technical, they’re organizational. If teams don’t know when or how to use AI, even the best models will sit idle or create confusion. A recent MIT study shows that 95% of AI projects fail, because leadership hasn’t set up clear strategies or plan on how to handle and use them.
Mindset shift: Move from AI-as-a-tool to AI-as-a-teammate. Treat it as part of the team: assign responsibilities, hold it accountable, and adapt workflows around it. Let it write a brief summary of it’s daily and weekly tasks in each morning huddle and manager check-in
The future doesn’t need to be feared. It needs to be trained.
How will we build the next generation of intelligent organizations?
|
