Internal Audit Conference: spotlight on AI with Priya Lakhani

“I want to show you the whole world of AI. Once you see that, you will realise that your organisation has already missed 30-40 years of opportunities,” says Lakhani .

AI is not an “it” and every organisation will have 50 to 100 AI applications that could transform every part of their businesses. But to identify these, internal audit needs to understand how AI operates “beneath the bonnet”. This doesn’t mean knowing how to programme your own AI system, but you need to look beyond the technological jargon and vague references to “AI solutions” and get to grips with the basic nuts and bolts. 

Priya Lakhani, lawyer, government adviser and technology entrepreneur, knows this from her own experience building an AI platform for schools. She will use her keynote speech at this year’s Internal Audit Conference to make sure that every member of the audience leaves understanding how to spot multiple opportunities presented by AI and discuss these, along with the risks, at every level in their organisation.

“I am not going to tell you how to write a prompt for a chatbot,” she warns. “I want everyone to leave the session with the core components of AI, so that no matter how the tech changes, your understanding is future-proof and you will always be able to spot the opportunities for yourself.” 

Take control 

This session will not tell you how others are using machine learning, or suggest solutions that you could adopt. It will not tell you what you should be doing or promise that buying specific software will solve a particular issue. But it will empower you to think about where it could be used across your organisation. You will leave with a list of questions that will enable you to create a high-level framework to deliver an AI strategy as soon as you return to the office.

“It’s all about data – you can’t have an AI strategy without a data strategy, and this is where internal audit is incredibly powerful,” Lakhani says. “You know where all the balls are in the business and how they interact with each other. This is critical, because it’s where the value is.”

Every use of AI should come down to outputs, she emphasises. No one else can tell you how to use AI in your organisation, because you need a pragmatic understanding of your needs and your resources. Internal audit is ideally placed to identify where machine learning can be best deployed to achieve results and offer support to the board about establishing the culture, opportunities and controls necessary.

What’s more, understanding how the technology functions will enable you to predict potential future disruption created by evolving AI and prepare for its effects. It might even enable your organisation to introduce disruption and become a disruptor. 

Innovative cultures and existential threats 

Her experience as a government adviser and in education means that Lakhani has personal insights into how few people really understand machine learning (including those who talk confidently about it). She understands it from the inside out as a technician and as an entrepreneur, having built her own system as well as her company – she has algorithms pinned above her desk.  

This puts her in an ideal position as a teacher and communicator. She now co-hosts a BBC News programme AI Decoded, as well as advising politicians, educators and businesses. Her enthusiasm and energy are inspiring. “I love knowing why things do and don’t work,” she says. She has studied disruption theory and is passionate about creating innovative cultures. 

“AI is an existential issue for many large organisations and some sectors,” she says. “People often say they understand the issues, but most don’t understand how the technology works, so they can’t answer critical questions about what it means for them and what they can do about it.”

Audit leaders cannot go to boards and audit committees and tell them about the opportunities and risks of AI unless they really understand what AI is and can explain this, she insists. “Understanding how AI works may make you realise that your organisation is using only about 5% of the opportunities already available with the technology you have,” she adds.  

But you need to be precise about the outcomes you want and realistic about the disruption your organisation faces. This is why examples of AI interventions that have worked elsewhere is not always helpful. Those who understand how it operates and have a culture that encourages innovation will be able to implement applications at every level and respond to disruption as it appears on the horizon. 

The Internal Audit Conference takes place at London’s QEII Centre on 8-9 October and online tickets are also available. The full programme and tickets are available now.