“How is AI revolutionizing the world of Capital Advisory and how our team is embracing ChatGPT as a valuable tool”
ChatGPT, the next revolution in productivity?
Though we are no AI specialists, we can’t go around the AI frenzy that has been occurring in the markets following the launch of ChatGPT at the end of last year. An unraveled widespread adoption of OpenAI’s Generative Pre-trained Transformer (GPT), which is a type of large language model (LLM) neural network that can perform various natural language processing tasks such as answering questions, summarizing text and generating lines of code while using a technique called deep learning to produce output that looks like it is produced by a human, has raised questions amongst many of us on how it could impact our jobs and ultimately, our lives.
If you can imagine how otherworldly the predecessor of the pc: “the mainframe” sounds to the youngest members of our team, while some of our senior members actually worked with this type of machinery at the start of their career, there is no doubt the pc has revolutionized the world we live in and brought us enormous productivity gains over the last decades. Will these promising AI technologies revolutionize our world once more?
The impact of AI transformers on (y)our business
A general outlook
According to Goldman Sachs, the answer is “yes”, as many as 300 million jobs could become redundant as these AI generative tools will leapfrog humans in certain tasks. This automation, however, should not be feared. It could unshackle workers from mundane tasks and unleash greater labour productivity. A study by Goldman Sachs, published on April 5th, suggests that generative AI could grow global GDP by 7% in the next decade. An overview from The Economist indicating industries with the largest potential productivity gains below:
What is striking is that, unlike past breakthroughs in machine learning, it’s skilled and predominantly white-collar sectors that are most exposed. Industries which rely on programming and writing skills are most exposed where some studies show more than 50% potential productivity gains within these areas. With AI functioning as an army of trainees, ready to brainstorm, design, draft, analyse and code that has yet to become experts, it is however easy to see certain areas in business where the application could have a direct impact such as in marketing where AI GPTs can generate marketing and sales copy which are tailored to your clients, in research and analysis where AI technology can be used to research a wide range of topics including market trends, competitors’ services or other business-critical information or in content creation such as articles, blogs, videos and social media posts where an AI GPT can be leveraged to create well-structured and engaging content in less time but, as well in language translation and routine programming.
Promising developments
Though the above is not exhaustive, one could easily see the enormous potential and efficiency gains this technology could offer. Therefore it is no surprise that companies of all sizes have been engaged in a fierce competition over the past few months using AI that can generate, process and analyse text, code and images. Microsoft notably showed a preview of their “AI co-pilot” which can generate an initial design in the applications of Microsoft 365 like Word, Excel, PowerPoint and Outlook. AI can open up the arithmetic magic of Excel spreadsheet software to all users who are able to describe the calculations they want to make in simple text. Microsoft also showed how its AI can summarize emails and even virtual meetings as they happen live in the Teams meeting software. At the same time Google-owner Alphabet, touted AI features for Gmail and a “magic wand” to make its own word processor smart, create presentations, take notes in meetings and compose emails.
AI, consultancy and credit intermediation
With Bloomberg as well developing their own AI GPT, Microsoft investing another USD 11 billion in OpenAI and having office integration on its way soon, it would be naïve to think change would not be upon us. A striking example of potential efficiency gains in our industry is given by Peter Hinssen from De Tijd: as Peter was invited to the annual strategic conclave of one of the largest private equity players and where the CEO proudly showed his ‘private’ version of ChatGPT. The company had fed an AI tool with information about all its deals over the decades and using ChatGPT, all employees were able to produce documents and prepare research reports in no time. The company’s CEO said:
“Usually we would hire a McKinsey for that and it would cost us hundreds of thousands of dollars. Now we can achieve the same thing in minutes instead of weeks, for a fraction of the cost.”
The company clearly requested not to use the public version of GPT, but only the internal tool. This ensures that no internal information leaks out.
Though we are still far off a world where any business small and large could develop their own private AI transformer, however, it sets the scene and gives us some flavor. How does that translate into our world of credit intermediation? Productivity gains could undoubtedly be significant, as displayed above. The AI use case with the most transformational effect on credit intermediation is the assessment of creditworthiness of prospective borrowers for credit underwriting. AI bots can sift through tremendous amounts of data from all kinds of sources, which could vastly speed up credit scoring and risk management processes as well as increase the execution time of the KYC process. Banks are using AI-based systems to help make more informed, safer and profitable loan and credit decisions. Some banks are delving deeper into the world of AI by using their smart systems to help make investment decisions and support their investment banking research. Firms like Switzerland-based UBS and Netherlands-based ING are having AI systems scour the markets for untapped investment opportunities and inform their algorithmic trading systems. While humans are still in the loop with all these investment decisions, the AI systems are uncovering additional opportunities through better modelling and discovery. Our reading of the above is that it all increases decision-making speed which, from our end, we would only encourage.
In terms of work products within our business, these tools will probably be the most useful for the information memorandums and “teasers” used to market companies to potential financiers or buyers. Most of these documents have passages coming from other sources such as company materials and desk research. AI tools could reduce drudgery by setting up the basic document and letting us, the intermediaries focus on how to “sell” the company or financing project. Though you could see compliance issues arise if the AI generated content in a IM or teaser is based on leaked documents or random comments on Reddit. As it concerns us, we are integrating carefully AI technology into for example when it concerns writing copy, creating content, generating fundraising ideas and performing desk research and look forward to capabilities of for example the Microsoft AI co-pilot to increase the efficiency and speed of execution of our daily work.
Despite the clear benefits of leveraging Artificial Intelligence in consulting, with its speed and efficiency, human consultants are still invaluable. Artificial Intelligence in consulting should be seen as an opportunity to increase human capabilities instead of taking them away. For example, Artificial Intelligence can crunch data faster than anyone and present trends and insights that would take humans days or weeks to discover. This newly freed time can then be spent thinking up creative strategies, analysing risk profiles and giving clients a better view on their opportunities. Put simply, if you’re looking for passionate strategizing and insightful problem solving, you’ll need human professionals – not Artificial Intelligence – behind the wheel. Artificial Intelligence may simplify tasks and allow us to work faster but it’s no replacement for a consultant who won’t rest until they have helped their client get the best outcome possible. That’s why we view AI as an ally – not an enemy – so that both monovalent forces can collaborate towards achieving outstanding results on behalf of clients everywhere. Overall, it is still a relationship-driven, client-facing business build on trust and reputation in which clients pay for our full and human attention.
Are we going too fast? – A closing note
ChatGPT had reached c. 100 million active users just two months after launch, making it the fastest-growing consumer application in history, this creates a first-mover advantage against other AI companies such as the LaMDA (BARD) from Alphabet and the European BLOOM initiative. Whereas the European initiative, to which over a thousand scientists contributed, carefully selected all data that is feeding the transformer other alternatives are using data from all over the web which might induce a number of problems.
The possibilities to maximize internal knowledge with these kinds of tools are phenomenal, but nobody wants to see their confidential knowledge just pop up in the next version of ChatGPT. Information architecture therefore becomes one of the greatest challenges of technology as many businesses may lack the IT architecture or inclination to accommodate AI innovations. In addition, the “creative” output the chatbots generate is based on a mashup of data sourced from the internet, raising issues around accuracy, privacy and intellectual property and human qualities, such as empathy, charisma and contextual reasoning, will be overlooked by AI bots. So for all their conversational panache, in the real world, AI tools will still need handlers. And that, may even end up creating new jobs!
Annex
As companies all over the world are keen to integrate the GPT-4 technology into their systems and you might be one of them, there are a few thoughts that we would like to share:
- Since the platform feeds on data across the worldwide web that is produced , while putting a layer of its own “human” interpretation on top, it can poor out prejudices, wrong and even made-up content. With the most recent technology, GPT-4, providing you with links to where the information is sourced, the issue however remains. OpenAI states that with this latest version you would get a 40% higher chance of getting a correct answer, though that doesn’t say much. The actual % of how much of the output is actually correct is nowhere to be found, indicating that this number might be substantially high.
- In the case of ChatGPT: OpenAI is a US-based company where personal data is treated differently compared to within the EU. Italy’s privacy watchdog has already banned ChatGPT over concerns about a data breach and the legal basis for using personal data to train the popular chatbot.
- The output of the model can be guided by its human creators to meet whatever ends they desire, with the danger being that its omnipresence and its human-like abilities have the power to create a convincing reality and trust where none does and should exist.
- While the technology is getting smarter, businesses should still validate the research.
- It is important to note that simply copying and pasting ChatGPT output directly into your company blog is technically plagiarism.
- Since ChatGPT leverages the internet for information, you run the risk of posting something that is too similar to its source, which can hurt your search engine rankings.
Sources:
https://www.theguardian.com/technology/2023/apr/03/the-danger-of-blindly-embracing-the-rise-of-ai
https://fd.nl/bedrijfsleven/14736.32/dagkoers-hoe-europa-de-strijd-om-de-chatbot-verloor
https://www.tijd.be/opinie/column/epo-voor-de-kenniswerker/10460072.html
https://podcast.standaard.be/episode/59059246
https://podcast.standaard.be/episode/56301965
https://www.linkedin.com/pulse/how-ai-machine-learning-revolutionising-lending-landscape/
https://www.tijd.be/opinie/column/before-chatgpt/10466280.html
