Is there “too much spend and too little benefit” in the AI ​​craze? By

Is there “too much spend and too little benefit” in the AI ​​craze? By
Is there “too much spend and too little benefit” in the AI ​​craze? By – Tech giants and other companies will spend around $1 trillion in the coming years to develop their artificial intelligence capabilities, including investments in data centers, chips and other AI-related infrastructure, according to Goldman Sachs analysts.

However, they argue that these expenditures have so far not brought much “other than reports of efficiency gains” among AI developers, while Nvidia (NASDAQ:) – Wall Street darling and the center of the hype surrounding the young technology – has seen a “sharp correction” in its shares.

To find out whether high corporate spending on AI brings meaningful “benefits and returns,” the investment bank spoke to a number of experts, including Daron Acemoglu, a professor at the Massachusetts Institute of Technology who specializes in economics.

Potential impact of AI spending on productivity

Acemoglu was largely skeptical about the outcome of the capital rush, estimating that only a quarter of all AI-related activities could be “cost-effectively automated” within the next ten years. This means that AI will take over less than 5 percent of all tasks.

“Over this (10-year) horizon, AI technology (…) will primarily increase the efficiency of existing production processes by automating certain tasks or increasing the productivity of the workers who perform those tasks,” Acemoglu told Goldman Sachs. “So the assessment of productivity and growth gains from AI technology over a shorter horizon depends entirely on the number of production processes that the technology will impact and the extent to which that technology increases productivity or reduces costs over that period.”

However, he predicted that there won’t be a “massive” number of tasks impacted by AI in the near future, adding that most jobs humans currently perform – such as manufacturing or mining – “are multi-faceted and require interaction with the real world.” Instead, Acemoglu expects AI to have the biggest impact on “purely mental tasks” in the coming years, adding that while the number of such jobs will be “non-trivial,” it won’t be “huge.”

Ultimately, Acemoglu predicted that AI would only increase productivity in the U.S. by 0.5 percent over the next decade, but would boost overall economic growth by 0.9 percent.

The “limits” of AI

Acemoglu added that he was “less convinced” that plans by major technology companies to dramatically increase the amount of data and computing power they feed into AI models would lead to faster improvements in these systems.

“If we incorporate twice as much data from the social media platform Reddit into the next version of OpenAI’s chatbot (ChatGPT), it might improve its ability to predict the next word in informal conversations. But it won’t necessarily improve a customer service agent’s ability to help a customer troubleshoot issues with their video service,” he said.

The quality of the data is also crucial, Acemoglu noted, pointing out that it is still unclear what the most important sources of high-quality information would be and whether it would be “easy and cheap” to obtain.

Finally, he warned that the current architecture of AI technology itself “may have limitations.”

“Human cognition involves many types of cognitive processes, sensory inputs and reasoning skills. Large language models (LLMs) today have proven to be more impressive than many people expected, but it still takes a huge leap of faith to believe that the architecture for predicting the next word in a sentence will achieve capabilities as intelligent as HAL 9000 in ‘2001: A Space Odyssey,'” Acemoglu said, referring to the fictional artificial intelligence character in a popular 1960s science fiction film.

An AI “bubble” or a “promising” spending cycle?

Goldman Sachs analysts see a mixed approach to the AI ​​spending slump, with some saying the technology has yet to demonstrate its ability to solve the complex problems needed to justify the increased spending.

The researchers also said they do not expect AI costs to ever drop enough to make it affordable for companies to automate a large portion of tasks. In general, the AI ​​story that has led to an uptick in the benchmark so far this year is unlikely to continue, the researchers said.

Despite these concerns, other analysts at Goldman Sachs took a more optimistic stance, predicting that AI could lead to the automation of a quarter of all work. The current increase in capital spending, they argued, appears “more promising” than previous spending cycles because “established companies with low capital costs and huge distribution networks and customer bases are leading the charge.” They also forecast that U.S. productivity would improve by 9% over the next decade and economic activity would grow by a cumulative 6.1% thanks to advances in AI.

Overall, however, Goldman Sachs analysts concluded that “there is still room for AI to run, either because AI is starting to deliver on its promises or because bubbles take a long time to burst.”

Leave a Reply

Your email address will not be published. Required fields are marked *