Q] How can human creativity and AI inputs be reconciled in developing brand strategy?
It is about removing the friction from the creation of those goods. Steve Jobs had a good definition of innovation, he said that innovation is creativity that shapes ideas that will lead to some sort of value. Innovation is all about accelerating, amplifying that ideation process to the point where you identify the idea that’s going to lead to the maximum output for your clients. What AI can do is free up and remove the friction to allow you to make that process much more effective.
When it comes to pitches and brand strategies, AI can help to align the briefs with your workforce and your expertise. In terms of brand strategy, it is a very complex problem. What product should I develop? What market should I take them to?
Ultimately, these technologies allow you to not only solve these problems that I’ve just described to you, but also remove friction, which could then allow you to create what are called digital twins. A digital twin of an organisation can run simulations, and then you can see how that simulation will affect that company. So, you have to ask, what organisations are moving beyond Digital transformation to these digital twins to strategize holistically across everything that they do.
Q] How and when will this techology come to India?
I am not too sure, but my company, which was acquired by WPP two years ago, has over 10 years of experience in building, applying AI to solving those problems and building Digital twins to allow organisations to adapt quickly to a changing world. Remember, adaptation is synonymous with intelligence. The quicker we can adapt, the more intelligent we are. And of course, we’re not only doing that with our brands, but we’re also doing that inside WPP to enable us to be the most innovative and adaptive company that we can be.
It’s also very alarming with all the things that AI is doing. Sometimes I wonder, if ChatGPT reaches another level, and take away my job. I think in the next 10 years, we’re going to see an explosion of new opportunities to apply your time and creativity to contribute to the world. I think there’s a huge amount of friction inefficiency and ineffectiveness across our organisations.
Q] What are some of the most critical ethical pitfalls marketers should be aware of?
There are two challenges that we need to be facing. I advise governments on how to navigate AI and how to mitigate some of the risks in the short term and the long term. So, the short-term risks are building AI or using any technology that might overachieve its goal and by doing so, sometimes it could cause problems in other parts of the system.
The other challenge associated with AI is building AI systems that are essentially making decisions, and you don’t know how they’re making those decisions. The only difference between software and AI is that some AI systems are opaque in their decision-making process. Regulators can becomes transparent and explainable when technology is making a decision with a material impact on people’s lives. If it doesn’t have a material impact on people’s lives, then maybe it doesn’t need an explanation. When we build AI systems, we ensure that they’re always explainable – not to tick regulations boxes, but because they can help you understand and know things about the world that enable you to make better decisions.
In the short term, we need to ensure that we’re making and building explainable AI systems so that we’re mitigating some of the risks associated with bias. In the long term, these technologies can have macro impacts on creating a post-truth world. For example, we don’t know whether a piece of content is authentic or not. The industry is going to play a critical role in making sure that any content that’s being pushed out there is authenticated and can demonstrate provenance. You can demonstrate its history and where it’s coming from. I think this industry will be leading the way in mitigating risks associated with things like a post-truth world.
Q] What are the future uses of generative AI, given that it is already being co-opted across media and technology?
Over the next few years, people will be able to access information and make better decisions based on reliable information, but of course, they will be able to remove friction from the supply chain and the production of goods and services. I’d like to think that we are at the forefront of applying generative AI. There are a lot of articles out there about WPP leading the way in terms of our relationship with our partners.
I believe we’ve already got a roadmap over the next five years about the implementation of these technologies to make us a better organisation.
Q] What are the most common mistakes made when approaching the use of AI and how can they be avoided?
There are so many mistakes. I think what we’ve seen over the past decade is companies not understanding what AI is or the nature of their problems. They’ve been doing what everybody else does – hiring machine learning experts or data scientists because they think that extracting insights from data will lead to better decisions and for the most part give human beings more insights. That doesn’t typically mean that they make better decisions.
Decision-making is a completely different field in computer science, which requires a completely different set of technologies, and skills. I’m a big advocate of making sure that we use the right algorithms to first solve a problem, and then work backwards to identify the right insights, and data to drive those insights.
I think some decisions should be made by AI, for example – there are five people in this room. If I want to allocate these five people to five jobs, there are 120 possible ways to do that. If I’ve got 15 people to allocate 15 jobs, I now have a trillion ways. If I have 60 people to allocate 60 jobs, I now have more possible combinations than there are atoms in the universe. AI can solve this problem significantly better than any human being.
There are thousands of decision-making problems that human beings are solving badly. If we apply AI, we can massively improve performance. But, companies are not yet open to using AI for solving problems or aren’t aware. Generative AI is good at knowing things about the world. It’s very good at telling you what it knows about the world through images and text. However, it is not good at making decisions on our behalf. We see companies making this mistake with regard to AI in a similar manner. They don’t budget for supporting those systems or for maintaining them, instead they hire expensive people to build these AI systems. What happens is that an expensive person ends up supporting some five AI systems, and they don’t get to do more interesting things.
Q] How new is this concept of AI being used for decision-making?
AI for decision-making has been used since Alan Turing broke the Enigma code. That was a complex optimisation problem. They used an algorithm to solve it, and no human would have been able to solve that problem. Since the birth of algorithms, we’ve been using them to be able to make better decisions, to be able to extract insights from data. I guess only in the past two decades, algorithms, cloud and data converged to allow us to do that exponentially quicker. But these technologies have been around for a long time.
Q] What are the key ethical considerations around AI implementation? Are there any existing regulations in place to address these concerns?
I have a relatively controversial view that there’s no such thing as AI ethics. There are a lot of people out there that are calling themselves ‘AI ethicists.’ Ethics is the study of right and wrong. But the difference between a human being and AI is that a human being creates the intent. The intent is to route my vehicles to maximise deliveries or allocate my workforce to improve well-being. They then build a system, whether it is an AI system or a people-based one to achieve that intent. So it is intent that needs to be scrutinised. From an ethics perspective, the AI is just doing what it’s told and where it gets it wrong, where it’s biased, or where it over-achieves its intent and causes harm, these are safety problems.
We need to make sure that we decouple these two things. Most of the challenges we’re dealing with from an AI perspective are safety problems in building software that behaves how we want it to behave. However, there are macro impacts that these technologies could have on society around building super-intelligence, post-truth worlds, and surveillance states.
The government is thinking about it, and WPP is also thinking about the UK. I’m from the UK and so I was recently an expert witness to the UK-published AI White Paper, and I commented on that paper in order to improve it. I believe the government has a critical role to play in creating regulations. Companies like WPP have a more important role in helping governments understand how these technologies could be applied, what the risks are, and how to ensure that we are creating the right guardrails, but not stifling innovation.
Q] Looking into the future, what are the expected long-term effects of AI on society, and how it might shape various aspects of our daily lives?
I am incredibly hopeful that these technologies can remove friction from the creation and dissemination of food, healthcare, education and transport, bringing the cost of those goods down, making them more abundant for people. They can become economically free to contribute to humanity in ways they want. I believe we all have an innate desire to create and contribute to humanity in ways that make other people’s lives better.
I would like to see AI over the next ten years, and I think it will free people from economic constraints, giving them the tools and power to make a virtuous, positive circle for humanity.
Q] The metaverse has gained immense attention. How does AI play a role in the development and functioning of the metaverse?
The metaverse for me has two important elements. One is the virtual environments and worlds where we are navigating and enriching our experience, how we are engaging with education and brands, and other things. So that will also open up a whole world of opportunities for people to create. Second, and what’s interesting for me is the Digital representation of you that exists across the metaverse, logging us into these different universes. Digital representation is gathering a lot of information about you, who you are, and your preferences.
I think we can augment you with your digital self to help you make better decisions about your life, career and health. So, you have an augmented AI that helps make you a better person.
