Two opposite signs of artificial intelligence; TSMC's boom and uncertainty about corporate demand

Source link: Two contrasting signs of artificial intelligence; TSMC's boom and doubts about corporate demand at a glance TSMC reported strong profit growth and very strong demand for artificial intelligence, saying supply still lags behind demand. Barclays downgraded shares of Dell, HPE and NetApp, arguing that corporate spending on artificial intelligence has yet to accelerate. These two messages show the growth of intelligence […] Source site: Safir Arjan

TSMC reported strong profit growth and very strong demand for AI, saying supply continues to lag behind demand.

Barclays downgraded shares of Dell, HPE and NetApp, arguing that corporate spending on artificial intelligence has yet to accelerate.

These two messages show that the AI boom is real, but for now it's mostly limited to big cloud and AI companies.

Two major tech news stories this week had seemingly conflicting messages about artificial intelligence. But together they show that the sector's boom is real, albeit perhaps more limited in scope than some valuations suggest. Demand is mostly concentrated among the largest cloud and AI companies and is not yet spread evenly across enterprise technology budgets.

The optimistic sign came from Taiwan Semiconductor Manufacturing Corp. TSMC, the world's largest contract chip maker. The company's profit rose nearly 70 percent in the third quarter, and TSMC raised its annual growth forecast. The company's CEO, CCV, said demand for artificial intelligence remains very strong and shows no signs of slowing down. The high-performance computing division, which includes artificial intelligence accelerators, now accounts for more than two-thirds of TSMC's revenue. The company says supply is still far short of demand.

A more cautious signal came a day earlier from Barclays. In downgrading Dell shares, the bank argued that companies' spending on artificial intelligence has not yet shown a significant increase. According to Barclays, a significant part of the revenue growth of hardware companies comes from increasing the price of memory chips and passing these costs on to customers. The bank made the same assessment about Hewlett Packard Enterprise and NetApp, and the drop in stock prices spread to other AI hardware companies.

These two views are not so contradictory. TSMC's customers are primarily chip designers who sell to the largest cloud service providers and artificial intelligence developers; Companies that continue to spend heavily. In contrast, Dell and its peers rely more on conventional businesses that are building their AI capacity. Barclays believes that the gap between expectations and reality is more obvious in this sector.

This difference is important for investors. If the demand for AI remains concentrated at the level of large companies, the main suppliers of these companies will have a more stable position than the businesses that depend on the widespread use of AI among companies. Also, part of the recent growth in enterprise hardware revenue is due to higher pricing rather than increased demand; A trend that may reverse if memory chip prices stop growing.

However, both sides of the story face ambiguity. Concentrated demand depends on continued spending by a limited number of very large buyers. On the other hand, adoption of AI among enterprises may accelerate as businesses move through the testing phase and move toward implementation. None of these trends have been finalized yet.

In order to clarify the picture, several signs should be followed:

Big Cloud Companies Spending Plans: Quarterly financial reports coming out from the biggest cloud service providers show whether their AI budgets are still on the rise.

Memory chip prices: The halt in growth in these prices reveals how much of hardware revenue growth is actually driven by demand.

Enterprise orders: The hardware manufacturers' own results will show whether AI server sales have expanded beyond a limited number of large customers.