AI could add $2.52 trillion to the global economy annually, says McKinsey. However, rising costs might cut this value by $850 billion. Companies plan to spend $7 trillion on AI infrastructure by 2030. Financial services and education show the biggest potential gains, but success depends on managing these high costs.
The McKinsey report highlights that a 30% increase in the effective cost of AI could cut the global economic value pool from $2.52 trillion to $1.67 trillion.
This would be a decline of around 34%, or $850 billion.
In the US alone, the annual value pool could drop by an estimated $320 billion under similar circumstances.
As companies pour money into AI infrastructure, software, and automation, questions are being raised about whether these investments can really boost productivity enough to justify the costs.
McKinsey estimates that around $7 trillion could be invested in AI infrastructure by 2030.
To justify this investment, there would need to be about $760 billion in annual profit pools.
The McKinsey report identifies financial services as the sector with the biggest potential economic gains from AI, at an estimated $530 billion annually.
Education follows closely with a projected boost of $410 billion.
Other sectors like technology, media & telecommunications, and healthcare are also expected to see significant benefits from AI adoption.
The report also highlights that the economic benefits of AI differ from country to country, depending on labor costs.
This is particularly important for labor-intensive economies where lower wages can alter the financial case for replacing or supporting human tasks with AI.
The study compares existing labor costs with the full cost of using AI, including technology expenses, human supervision, verification, and rework.