Raging debate over the pros and cons of AI

Ruling the world: Signage at the Nvidia Corp offices in Taipei, Taiwan. In 2023, close to US$315bil was invested in tech companies globally. — Bloomberg

THE world is in a mess, and the acceleration of artificial Intelligence (AI) usage is disrupting every business and the way we live.

We are all struggling to understand what the AI transition means for each of us as consumers, parents, teachers, businesses or government leaders.

The debate over the pros and cons of AI is raging, especially in its dual military-civilian usage. AI will guide the next drone or missile at you with faster accuracy than ever imagined.

It can also develop the next miracle drug to change our health. We simply do not know what the outcome whether AI is ultimately good or bad – only that the bandwidth of risk and opportunity is widening at frightening speed.

We have never seen technology being adopted in daily activities in terms of speed, scale and scope as AI. The AI revolution has pushed the Nvidia and other AI platform stock valuations into the trillion-dollar league.

Big powers and big platforms are all investing in AI, trying to figure out how to beat the competitors in achieving scale and domination.

The digital divide means that those who are ahead in AI will be richer, faster, smarter and more powerful, whereas those who don’t implement AI tools are being marginalised.

Clearly, the rich and advanced economies stand to gain more from AI and technology, whereas emerging and developing market economies are still struggling on how to use AI to help them develop or at the minimum, tackle their myriad problems of people and planetary injustices.

The most obvious benefit of AI is that it could improve productivity, which has declined globally across the board for several decades.

McKinsey research suggests human-centric generative AI adoption may well automate up to 30% of business activities across occupations by 2030.

Analysing 63 user cases, they estimated that generative AI could add roughly US$2.6 trillion to US$4.4 trillion annually to the global economy, equivalent to adding 2.5% to 4.2% to current global gross domestic product, which has been forecast by the World Bank to slow down to half the growth before the global crisis in 2008.

The potential for turning around development in multi-directions using AI looks huge. How can this be achieved?

AI is essentially a human-invented tool for learning and using for change.

Given the right amount of data, it can help make better decisions and eliminate inefficiencies in the system. It can also do bad things at scale.

Ethics in the usage of AI is at the heart of the current debate. In the wrong hands, AI is what historian Yuval Noah Harari calls “data colonisation and digital dictatorship”.

Nobel Laureate economist Joseph Stiglitz propounded that the job of governments was to create a learning (knowledge) society, since knowledge is a public good.

Fellow Nobel Laureate Robert Solow (1924-2023) first quantitatively identified that the most important determinant of economic growth was technological change.

Kenneth Arrow (1921-2017) showed that markets by themselves do not yield efficiency in the production and dissemination of knowledge.

More recent case studies on building tech ecosystems showed that learning is really about copying or imitating global knowledge and adapting these to local needs.

South Korean Professors Kim and Lee (2022) showed that Taipei and Shenzhen evolved into tech powerhouses by first importing foreign technology through welcoming multinational companies and then developing local champions that increased research and development, primarily in process engineering, and then moving to original ideas, products and services that began to rival foreign competitors.

In short, human learning is always about copying others and then personalising or internalising such knowledge to create new ideas and actions.

This “copy-learn-adapt-innovate-scale” approach is exactly the path that AI usage is following.

When we face something totally new, we have four essential choices. The first is to deny or reject because we fear the unknown. The second for those who are curious is to learn and experiment. The third is to do nothing or simply follow the crowd, because that appears to be the safest way out of disruptive change. The brave and risk-takers are those who decide to leap into the unknown and become innovative or entrepreneurs. These become the change agents.

In today’s existential threats of nuclear war, ecological collapse and technological disruption, doing nothing or business-as-usual is not an option. You either eat lunch or become lunch.

There is no complacency shown in the financial sector, with the Bank of International Settlements and Citigroup Inc recently warning about the profound impact and opportunities and threats of AI on the financial landscape.

The amount of startups working on implementing AI in different domains is staggering. In 2023, close to US$315bil was invested in tech companies globally, a large chunk being in AI applications.

The World Economic Forum (WEF) has been promoting application of AI in social innovation to tackle social and ecological issues. Although the WEF report stressed that “Artificial intelligence has the potential to scale impact in several domains but requires collaboration to help social innovators realise its maximum potential”, the greatest barriers to successful AI implementation are lack of trust, partnerships and funding.

The pattern in adopting AI in every domain, from personal to communities, businesses and government is common. You must approach change from a complex system perspective, noting that there are no simple one-size-fit-all solutions.

Change management is not rocket science - it is about changing mindsets, addressing vested interests, and having the passion and management skills to execute change.

An excellent Japanese study on regulation shows that 20% of staff time are spent on compliance issues, so that small and medium enterprises face overwhelming costs and barriers to deal with daily regulatory and bureaucratic issues. Just simplifying overlapping silos in bureaucracies using AI tools could increase productivity by 8%.

In sum, we all need to adopt AI tools to generate the productivity that is needed to achieve more with less. Although change is best tackled bottom-up, it needs leadership, courage and passion to engineer change. That takes human intelligence, with AI as a tool, but impactful change is never about one individual, but about the whole and all of us.

Andrew Sheng writes on global issues from an Asian perspective. The views expressed here are the writer’s own.