There’s a lot of noise about AI in HR right now – from chatbots in recruitment to AI literacy training and compliance with the EU AI Act. So we brought together a room of senior HR leaders, over good food and better conversation, to unpack the issues, the realities on the ground and where things are heading. 

These are 5 of the ideas that came out of our discussions:

1. The promise of AI in HR is clear. The execution? Still early.

Everyone sees the promise of AI in HR. But when it comes to real implementation, most organisations are still at the experimentation stage. A few are piloting use cases or tools, but scaled, consistent adoption is rare. Time is often the blocker. With so much else going on in HR, thinking strategically about AI can get pushed down the list -  and that’s where organisations risk falling behind.

Suggested Resources: 

• Follow our AI expert speaker Mark Breslin to keep up with his work across multiple organisations adopting AI. 

• For inspiration on where AI can add most value at your company, check out some examples of how companies are using ChatGPT

2. The EU AI Act Is Here But The Details Are Still Fuzzy

The EU AI Act is landing just as EU employers are still digesting a range of other new compliance obligations. 

And while the EU AI Act might feel more like the domain of the CIO or CTO, HR is actually very much in the spotlight as the Act specifically calls out AI systems used in hiring, promotion, and workforce management. “High-risk” was a term that came up a lot — especially around recruitment tools.

Also, the Act calls for AI literacy training for employees which will typically be under the remit of HR. Most haven’t started. Now’s the time.

Suggested Resource: The EU AI Act: The Download for Employers by law firm, Matheson

3. Having an AI Strategy Is Not Enough – You Need Enablement

Having an AI strategy is a great start. But it’s not enough on its own. The companies making real progress aren’t just issuing top-down mandates — they’re also creating space for bottom-up experimentation.

Leadership defines ambition, tone, and ethics. But the real innovation happens at the edges — when teams closest to the work get to test, learn, and apply AI in their own context.

Suggested Resource:  The Innovation Show Podcast with Aidan McCullen, Don’t Let Hierarchy Stifle Innovation with Timothy R. Clarke

4. The Real Risk? AI Usage Without Skills or Oversight

Some companies are locking down LLMs out of fear. But that often drives employees to use unapproved tools anyway — with less control, and more risk. Think: personal identifiable information or company IP pasted into public models. 

A better approach than ‘shadow AI’ is to create a safe way to use AI responsibly. That means policies, guardrails, and training — not bans.

Suggested Resource: How To Implement AI — Responsibly by Michael Wade and Tomoko Yokoi, Harvard Business Review

5. AI Literacy Starts With Each Of  Us

AI literacy isn't just a compliance checkbox - it's a leadership capability now. If we’re not up to speed, how can our teams be? Understanding what AI is (and isn’t), where it fits, and how to use it responsibly is now part of the HR toolkit. 

Suggested Resources: 

Follow Allie K. Miller or Ethan Mollick for bite-sized, practical takes on how AI’s showing up in work or try a course or newsletter from DeepLearning.ai  

Final word:
We all left the dinner with more clarity, new connections, and a sense that no one has this fully figured out yet -  while though ChatGPT launched almost 3 years ago, enterprise adoption of LLMs and other AI tools is still early. The good news? You don’t need to have it all figured out. But you do need to start.

The next SkillsTrust Dinner is coming soon. If you’d like an invitation or to suggest a topic, just let us know: !