What is ChatGPT?
In November 2022, OpenAI launched ChatGPT, a chatbot (i.e. a software application), with which humans can have a conversation and that answers to most of users’ queries. The ability to generate content in a human-like fashion is what distinguishes it from previously developed AI tools, which are mostly used to make predictions based on our behaviours. Furthermore, it is the reason why this tool is classified as generative AI and why it might signal the beginning of a new era for AI.
Why is it important?
ChatGPT was a global phenomenon and reached 100 million users in just over two months, making it the fastest growing internet service ever. But more importantly, many industry leaders believe generative AI could be one of the most important advances in technologies and compare it to the creation of the web or the advent of the first iPhone. What is becoming clear is that generative AI could be a boost to productivity and could unlock creativity of people, thus radically changing how many industries work. For example, an AI solution that recommends strings of software codes is already helping developers to do their job in almost half the time that it usually takes. Furthermore, according to PwC, AI could add up to 1% per year to global GDP growth for a total value of USD 16 trillion by 2030.
Can ChatGPT disrupt the search market?
Microsoft, which owns an interest in OpenAI, announced that they will integrate ChatGPT’s technology in their search engine. Their declared target is to take market shares in the lucrative search market. Despite their intents, we do not believe that the search market (currently dominated by Google with a market share of 92%) will see major market-share shifts for now due to: the behavioural changes needed to change the status quo, the low monetisation potential of open queries and their higher costs, and the fact that Google is a quality player with plenty of AI capabilities to defend its position. From a financial perspective, we believe that higher competition within the search market will not jeopardise the revenue base, but we see potential for further monetisation opportunities. Furthermore, additional offerings and cost-saving initiatives might help to offset higher operating costs and investments into AI capabilities.
How can ChatGPT be commercialised?
Adding AI capabilities to existing products and creating new ones are exciting opportunities for Microsoft and other software companies. This will lead to a premiumisation opportunity, which should drive further growth for software companies. For example, Microsoft launched Teams Premium, available at a monthly fee of USD 10 per user, which leverages AI capabilities to automatically summarise video meeting discussions without the need of a human assistant. However, for software companies, AI is an opportunity as much as a threat as it might create more competition within certain sub-segments. The key to success is to own proprietary data that can be used to fine-tune AI models and provide better products to customers, thus justifying price increases for premium products.
How can ChatGPT benefit intermediaries?
ChatGPT can offer intermediaries the ability to analyse complex financial data and provide insights into potential investment opportunities, allowing them to make more informed decisions. And with its natural language processing abilities, it can also assist intermediaries in conducting research and monitoring market trends, saving them time and resources. Additionally, ChatGPT can assist in risk management by providing real-time alerts and suggesting strategies to mitigate potential risks.
What other sectors might benefit?
There are other sectors that are well positioned to capture a significant portion of the value created by AI solutions, namely semiconductors and the semiconductor value chain, cloud infrastructure, and other hardware producers.
What are the risks?
AI poses sustainability challenges in relation to its impact on the job market and on the environment. However, the reputational and regulatory risks associated with misinformation and the harmful use of AI are potentially the most material risks for AI companies. Indeed, it is important to note that AI and natural-language models are prone to making errors and generating misinformation, while they might infringe data privacy, security, and intellectual property.
What is our view of generative AI?
We are excited about generative AI’s potential to unlock productivity and creativity. We see generative AI as having a personal assistant that helps us 24/7, both in the workplace and in our personal life, thus giving us more time and resources to be used for what we feel is most needed. We believe we are just scratching the surface of what generative AI is capable of and expect most industries to gradually integrate it into their processes as well as find new and original applications. Adopting this technology in an efficient and safe way will be a trial-and-error process. Risks shouldn’t be underestimated - but the potential benefits are immense.