As it is trained, generative artificial intelligence (“gAI”) ingests and processes vast amounts of data. It then uses the acquired knowledge and discovered patterns to create new content. Thanks to its cognitive, predictive, and language abilities, gAI is an invention that invents while reinventing itself.
As such, gAI has the potential to overcome the adverse innovation drivers discussed in ‘The Burden of Knowledge’ and fuel a technological renaissance. McKinsey estimates that gAI could add more than $4 trillion annually to the global economy (on top of traditional AI). That would represent a 4% uplift to global GDP. Incidentally, this estimate is roughly equivalent to the annual global clean energy investments required to achieve Net Zero by 2050.
It is probable that this figure vastly underestimates gAI’s economic potential. In ‘Power and Predictions: The Disruptive Economics of AI’ (2022), the authors expect the use of AI to evolve from today’s ‘point solutions‘ and ‘application solutions’ (as in McKinsey’s study) to transformative ‘system solutions’ leading to a holistic redesigning of economic processes. Anyone concerned with a global supply squeeze (labor, natural resources, deglobalization) and lasting inflationary pressure need not be. If anything, deflation may become prevalent.
That said, gAI has triggered negatively charged emotions due to technological challenges (e.g., inaccuracies due to hallucinations), ethical questions (incl. misuse by social media or the military), and potential threats to the human order. Radiohead’s magnificent ‘OK Computer’ (1997) plays loudly in the background.
There are reasons to be concerned. The more gAI machines are trained, the more effective they are. Hence, there is a race down a steep learning curve for every economic actor, especially those in knowledge-based industries. FOMO is a powerful competitive driver, but it is already generating disturbing anxiety across society, considering gAI projects’ both intended and unknown consequences.
Moreover, gAI may trigger a human identity crisis. Discovering new, human logic-defying, and inconvenient scientific truths through traditional AI is already shaking established belief systems. But gAI will raise profound philosophical questions about what it means to be human, the nature of creativity, and the role of technology in society.
Finally, through faked intimacy (see the movie ‘MƎGAN’ (2022)), this ‘alien intelligence’ can exploit individuals’ psychological weaknesses and offer alternative, seductive truths influencing voting or purchasing behavior. Yuval Harari argues that the gAI threat to humanity is not physical but mental.
Aside from these considerations requiring tight risk management to promote ‘beneficial gAI,’ gAI will revolutionize corporate finance (e.g., Bloomberg-GPT) and deal-making. New tools will soon support investment target screening in the public and private markets, due diligence (e.g., M&A data room analysis), valuation based on an exploding volume of pre-financial (ESG) data, and contract drafting.
Since gAI will shorten due diligence processes and reduce transaction risk, M&A activity will rise and unlock significant economic benefits through optimized capital allocation – beyond McKinsey’s estimates. That is at least one thing worth looking forward to.
Sources - Books
Power and Predictions: The Disruptive Economics of AI’ by Agrawal/Gans/Goldfarb (2022)
‘The Age of AI’ by Kissinger/Schmidt/Huttenlocher (2022)
‘Human Compatible: AI and the Problem of Control’ by Stuart Russell (2020)
‘Life 3.0’ by Max Tegmark (2017)
Sources - Articles
‘Paradigm shift in sustainability disclosure analysis: Empowering stakeholders with CHATREPORT’ by various authors (2023)
‘Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models’ by Lopez-Lira/Tang (2023)
‘Behavioral Machine Learning? Computer Predictions of Corporate Earnings also Overreact’ by Frank/Gao/Yang (2023)
‘Generative AI-nxiety’ by Reid Blackman in the Harvard Business Review (2023)
‘The state of AI in 2023: Generative AI breakout year’ by McKinsey (2023)
‘How to train generative AI using your company data’ by Davenport/Alavi in the Harvard Business Review (2023)
‘The economic potential of generative AI’ by McKinsey (2023)
‘AI needs to progress to jumpstart our economic productivity’ by Steve Rattner (2023)