When Chat GPT came on the scene, the chatter was about how this new technology was going to change everything. The truth is, as with most technology, AI and machine learning didn’t suddenly arrive - it was already being deployed in many industries in non-visible ways.

So, although there is no handbook, there is some precedent we can learn from. For example, financial services have long used forms of AI for fraud detection, algorithmic trading, and risk management. Retail has deployed it in inventory management and customer services (in some cases much to our irritation!) and in healthcare it is used reduce errors and lead times in diagnostic imaging.

What works in one context may not work in another, and this is a pretty complex and adaptive challenge for you as a leader, so the cardinal rule is to experiment, iterate and learn what the system you’re leading can do.

Here are a few additional pointers that may help you:

Get clear: Do you see AI as a means to reduce headcount or to enhance value?  Is it about freeing up resource or freeing up human energy for more creative work? Fuzzy messaging – or worse, silence – on this point risks engendering fear and resistance from your people.

DIY AI: Discover it yourself – explore and experiment and share what you are learning. This isn’t just for your learning benefit but also role-models healthy curiosity to the wider organisation.

Be the leader in the loop: Just as it is vital to always check the output of any large language model (LLM), it is crucial that business and functional leaders do not ‘outsource’ to technical leaders. How, when, and where to integrate AI into your operation are business strategy questions and you must take ownership of that.

Realistic optimism: Encourage curiosity and optimism about how AI can simplify the business and balance it with risk awareness (including ethical concerns) and limitations. 

Radical experimental learning: Provide the psychological and digital safety for people to invent use cases, and to share what works and what does not. 

• Talk about tasks that AI can help with, rather than about roles that AI can replace.
• Provide clear boundaries within which to welcome experimentation so that your data is protected and risks are mitigated.

Lead the team to include AI as a co-worker.  ​​​​

If you're interested in AI and its implications in our working lives, I highly recommend Co-Intelligence: Living and Working with AI by Ethan Mollick. AI is a tool with powerful capability, but we need the understanding and wherewithal to use it effectively and safely. 

For another dimension of leadership and AI, I recently wrote about leading for a culture of curiosity and safety in AI, and the importance of smart discourse - you can read it here.