Goldman Sachs, MIT Analysts Split on AI’s Future: Bubble or Breakthrough?

Artificial Intelligence (AI) has been the rage of recent times, a hot talking point across many fields, covering traditional finance and crypto – but experts are at odds over whether this is just a bubble or the next frontier in tech.

A report by Goldman Sachs and the Massachusetts Institute of Technology (MIT) didn’t reach a clear conclusion, with one Goldman Sachs economist and one MIT professor sceptical about AI’s transformative powers and three Goldman Sachs economists predicting a “killer app” to soon further boost the sector.

Related: Ancient Ethereum Whale Moves 7,000 ETH to Exchange Kraken

Underlying Technologies of Generative AI Inherently Complex

MIT’s Daron Acemoglu said while progress in generative AI will continue, the path to transformational change is more gradual and measured than it might seem.

Given the focus and architecture of generative AI technology today… truly transformative changes won’t happen quickly and few—if any—will likely occur within the next 10 years.

Daron Acemoglu

Acemoglu added that humans will remain “in the driver’s seat” even if current LLMs (large language models) achieve superintelligence – because they lack human capabilities.

It is very difficult to imagine that an LLM will have the same cognitive capabilities as humans to pose questions, develop solutions, then test those solutions and adopt them to new circumstances.

Daron Acemoglu

The extended timeline allows for a more thoughtful approach to development, aiming to maximise benefits while minimising potential risks and disruptions.

Goldman Sachs economist Jim Covello stated that while AI technology holds potential, its high development costs necessitate that it be effectively aligned with applications that can fully leverage its capabilities. Without this alignment, the justification for such investments becomes tenuous, particularly in scenarios where AI is not aptly suited to address the complex challenges it is tasked with.

AI technology is exceptionally expensive, and to justify those costs, the technology must be able to solve complex problems, which it isn’t designed to do.

Jim Covello

AI Has Ability To “Generate Returns Beyond The Current ‘Picks And Shovels’ Phase”

Fellow Goldman Sachs economists Joseph Briggs, Kash Rangan and Eric Sheridan hold a more positive view, believing in AI’s “ability to ultimately generate returns beyond the current ‘picks and shovels’ phase, even if AI’s ‘killer application’ has yet to emerge”.

Rangan admitted that the current level of spending in absolute dollar terms was high but he also said it was worth long-term

Related: Entertainment Giant Sony to Launch Crypto Exchange in Japan

Spending is certainly high today in absolute dollar terms. But this capex cycle seems more promising than even previous capex cycles.

Kash Rangan

While they don’t deliver any hint to what that killer app could be, they say there’s still “room for the AI theme to run”, they said. Ultimately, they believe AI will either “deliver on its promise” or slowly die, “because bubbles take a long time to burst”, as they put it.