Top take aways from Donald Farmer’s Keynote talk at the Business Change and Transformation Conference Europe 2024.

The Limitations of Current Generative AI Models

It is important to understand that generative AI models have limitations. AI depends on human prompts, and can display some bias. It works from existing images and produces approximates, where errors can creep in. The fold in a sleeve or a hand may look good at first glance, but on closer inspection may be completely wrong. We need to look at AI generated outputs with a sceptical eye and be aware of potential problems – bias, errors or hallucinations.

Is Generative AI Creative?

If we imagine a child playing with paint, it is a form of play. It is enjoyable. AI cannot experience the pleasure and excitement of creativity, where you fail and do better and enjoy the process.

“Show me a child who doesn’t want to hear a story and tell it back. Then we put them through a utility education and say that the arts are a luxury. The arts develop our natural creativity. If creativity is a luxury, then being human is a luxury.”  

– Jeanette Winterson.

If not Creative, AI can Spark Creativity

Donald Farmer’s artist wife Alison asked generative AI for ideas about crows, often seen as black and as harbingers of doom. The AI was able to inform her that to other birds, crows are colourful with distinctive ultraviolet patterns invisible to the human eye - AI was a creative partner to Alison.

Can Generative AI be Innovative?

Innovation is fresh work that emerges from radically new answers to the basic questions of our craft. The strength of AI is that it can keep on innovating. For example, when tasked with the design of a computer mouse generative AI will keep on exploring the latest materials, the latest programs and so on.

The Who? What? Why? When? Where? and How? of Innovation

You can innovate by changing the standard answers to one or two of these six questions. For example, if you answer the question ‘Who vacuums the floor?’ and decide to explore the answer ‘Nobody’ then you have the self-operating vacuum cleaner. You can prompt AI to be innovative by asking AI to answer some of these questions differently, to explore possibilities.

The Promise of Human-AI collaboration

Generative AI depends on good quality prompts. Defining a new strategy offers a good example. You must be able to describe your strategy to the AI, and the parameters, constraints and variables of the strategy. It can then structure an explanation of your strategy and simulate alternative strategies, conducting rapid prototyping and accomplishing in three minutes – this would take a team a few days to iterate. You can use this work as a starting point for developing a new strategy. The generative AI is your strategic advisor, or assistant.

AI can be used as an innovation accelerator to automate the ‘drudge’ work, freeing up humans to focus on creativity and ingenuity. Research shows that AI stimulates human innovation.

Key Recommendations

• Ensure you can describe prompts in ways AI can understand
• Practice with generative AI for innovation
• Learn the limitations through experiment and research
• Invest in human creativity

Donald Farmer is a visionary data and analytics strategist with over 30 years’ of experience designing data and analytics products. He has led teams at Microsoft and Qlik Technologies and is now the Principal of TreeHive Strategy where he advises software vendors, enterprises and investors on data and advanced analytics strategy.