AI still sets the pace in tech, though the winners are changing. Hardware drove the first leg of the rally, while megacap platforms are now drawing investors with stronger earnings and more ways to make money from AI. The largest tech stocks are also starting to move on their own catalysts again.
Megacaps regain control
The Magnificent Seven have returned to the center of the market after months of uneven performance. Fundstrat believes the Roundhill Magnificent Seven ETF is close to breaking out of an ascending triangle, a chart pattern that often signals further gains.
Alphabet, Amazon, Apple, Meta Platforms, Microsoft, Nvidia, and Tesla account for 31% of the S&P 500’s market value. Their size gives the group enough weight to pull broad indexes higher, while any stumble can quickly reach investors through passive funds.
Recent gains have come from different places. Apple improved after launching its foldable iPhone, Meta benefited from excitement around its Muse personal AI agent, and Microsoft steadied as fears about AI weakening software demand began to ease.
Tesla remains the group’s clear laggard. The other six do not need it to lead, but enough of them must keep rising for the ETF’s potential breakout to hold.
Valuations reset the debate
Tech’s pullback has changed the price of joining the AI trade. The sector now trades at 21 times earnings, down from 32 times in October, while forward earnings estimates have climbed 19% over the past three months.
Those numbers give investors a better entry point than they had during the most crowded stage of the rally. Tech valuations remain high, but earnings forecasts are rising while the multiple paid for them has fallen.
Training large models still consumes enormous amounts of money, with AI labs carrying most of the bill. Inference works differently because it powers the responses customers receive when they use AI products.
More inference means more paid searches, software features, advertising tools, and cloud usage. Megacap platforms already own those customer relationships, giving them a clearer route from AI spending to revenue.
Cloud platforms sit in a stronger position when AI spending follows customer usage. Amazon, Alphabet, and Microsoft will still need chips, data centers, and power, but they can recover those costs through recurring cloud and software revenue.
Truist analyst Sam Grelck expects slower model development to create uncertainty without necessarily reducing overall demand. Spending could simply move toward the services that put existing models to work.
Hardware loses its easy story
Concerns about the pace of AI development have put infrastructure stocks under pressure. Comments from Anthropic CEO Dario Amodei calling for a slowdown pushed South Korea’s KOSPI down 3.3%, while the Nasdaq lost 0.8%. Chipmakers absorbed the heaviest selling.
Their valuations rely on data-center construction and model training continuing at speed. Software companies have another source of growth because they can sell AI features through products customers already use.
Steve Sosnick of Interactive Brokers said other markets showed little sign of an approaching freeze in AI spending. Futures and bond yields would likely have come under much greater pressure if investors expected capital spending to collapse.
Dennis Dick of Triple D Trading described the selloff as money moving from hardware into software. Investors are becoming less willing to pay the highest multiples for suppliers that depend on constant capacity growth.
Nvidia remains the leading name in AI chips. Amazon, Alphabet, and Microsoft are closer to the customers paying for cloud capacity and software, which could give them an advantage if growth settles at a slower pace.
Labor data shows the shift
Hiring patterns tell a similar story. The tech sector shed about 14.7K jobs in August, even as companies across the wider economy added 86K technology roles.
Businesses are trimming general tech teams while hiring people with AI skills in finance, healthcare, retail, and other industries. AI is spreading into everyday operations as technology companies become more selective about headcount.
Broad tech funds still provide access to the megacap leaders, along with companies facing weaker demand and tighter margins. Platforms with established customers and distribution have more ways to earn back their AI spending than suppliers relying on nonstop infrastructure expansion.
Investors now need proof that AI spending can become durable earnings. The companies producing that proof are likely to lead the sector’s next move.