AI: the real change is not technology
The real change goes beyond technology. Businesses should not put 'the cart before the horse'. First, be clear on why you want a Gen AI solution and what it means for the company.
This past year has seen GenAI – and especially Chat GPT – coming into the mainstream.
I have decided to write about GenAI because most AI conversations seem to be too technology-centric. AI is not just about technology, it is about strategy, change, people and behaviours, and I know a lot about those. I have worked at the intersection of technology and organisational change for many years, often acting as the ‘translator’ between my tech colleagues and the rest of the company. This is an essential role: many people don’t read user guides, or remember training materials. People understand stories, purpose and meaning they can relate to. If you get these right, then you are on the path of success to introduce any new technology.
The alternative perspective on AI
I take a strategic and purpose-led perspective on AI. I challenge the implementation of any technology without first being clear about what problems it will solve. I advocate the priority of thinking first about strategy, benefits and impacts.
The key message is let’s not put the AI cart before the horse.
Keep reading if you want an alternative perspective on AI. There will be a new article every Tuesday and you can subscribe to receive it straight in your inbox. My measure of success is knowing that executives and business decision-makers will prioritise strategy, benefits, impacts and ethics before jumping into the AI cart.
Latest fad or here to stay?
There is an ongoing flow of emerging technologies. In the past years we had metaverse, Internet of Things, virtual/augmented reality, cryptocurrency, self-driving cars … creating a buzz and then moving aside for the next big thing. It is therefore legitimate to ask whether AI can be considered a fad or transformative, and therefore whether is worthy of investment. There are two important points to consider:
Artificial Intelligence is not a new idea, the term itself was coined and entered the lexicon with the publication of “Computer Machinery and Intelligence” by Alan Turing in 1950. He proposed a test of machine intelligence called The Imitation Game. Many of us are more familiar with the 2014 movie of that title with Benedict Cumberbatch. Since the 1950s AI has re-emerged cyclically in the mainstream.
The ‘generative’ adjective associated with AI seems to indicate a transformative technology with the potential for greater impact than other technologies, which are still important but haven’t had the expected ground-breaking impact.
In conclusion, the jury is still out and we don’t know for sure. We must remember that it’s only when we look back in time that we can accurately assess the true impact of anything.
Ethics and AI
More than with any other technology, ethics is a major consideration when we discuss AI. What strikes me most is its capability to learn. What scares me is the potential damage we can suffer if people trust it so much that they give up their agency, unknowingly follow fake news and it distorts decision-making.
AI is not just about technology
Back to the purpose of this article and the foundation of my writing here. Many companies tend to jump headfirst into new technology. It’s the ‘bandwagon effect’ in behavioural science, I also like to call it the ‘shining new toy’ effect. It’s a very normal reaction and we do the same (in reality or daydreaming) with fashion, cars, beach houses or holidays. The problem is that it can be an expensive mistake.
The challenge is that companies are complex and with many disparate components. I have learnt that introducing a new technology is never straightforward, it has an impact on the tech you already have in place. Systems often don’t ‘talk’ with each other, or require expensive workarounds to integrate.
And that’s just the technology part. I believe the organisational impacts are even more complex. There are impacts on the organisational structure, operations, and people. And of course, AI needs data: accurate, relevant and clean. And lots of it.
Let’s take an example, which is also the entry point of many companies’s AI journey: introducing am AI-powered chat assistant. Here are some of the requirements and impacts, basic ones applying to most organisations:
The chat assistant won’t be plug and play
Need accurate, relevant and clean data. Someone has to feed that data
May need to restructure departments/operations to reflect the presence of the AI assistant
Need to integrate it with other systems
If you have people answering customer questions, you will need to decide what to do with those humans
When I discuss new technology and organisational impacts with clients, we end up with a long list of potential impacts and a longer one of questions that need an answer.
This is why ‘getting AI’ is more than introducing a new technology and not something any executive should jump into headfirst.
We need a holistic approach, it starts with asking why you want it and what it can do for you and continues with an impact assessment. This is before you even reach out in the market to start talking with solution providers and data scientists.
In conclusion, my first advice is to look at your company’s goals and assess how GenAI can support them before you start spending money and resources. Subscribe to this stack to find out how.
Ps I created the image using DALL-E 3 via Bing. It’s my first time playing around with AI generated images. I kept asking for an image of a cart before the horse, but it didn’t come up with what I wanted… I will try out some of the other tools and write about it soon.
Share this post
Leave a comment