The biggest winners from artificial intelligence may not be the companies building the most popular AI products. According to Div Garg, founder and CEO of AGI Inc., the companies with the most to lose in an AI-agent future could be some of today's biggest technology platforms, including Alphabet Inc. (GOOGL) and Meta Platforms Inc. (META).
Garg, who famously turned down a lucrative job offer from OpenAI to build his own company, believes investors are focusing on the wrong question. Rather than asking which company develops the smartest AI model, he says the more important question is what happens when AI agents begin performing tasks currently handled by humans.
"The most vulnerable are those that rely heavily on human users," Garg said in an exclusive email interview with MarketDash.
That view strikes at the heart of some of Silicon Valley's most successful business models.
Why AI Agents Could Challenge Google's Search Business
For Google, the risk is straightforward. Search advertising depends on users conducting research, comparing options and clicking links. But if AI agents increasingly handle those tasks on behalf of consumers, the traditional search journey could become less relevant.
"Google won't be able to count on revenue from search, as people will no longer be doing the research themselves," Garg said.
Meta's Advertising Model Faces a New Test
Meta faces a similar challenge. The social media giant generates the overwhelming majority of its revenue from advertising targeted at human attention and engagement. In an environment where AI agents increasingly act as intermediaries between users and digital services, that model could face pressure.
"Meta, for instance, will need to find a revenue stream that doesn't rely on ads," Garg said.
The disruption may not stop with internet giants.
Uber and Airbnb 'Will Also Need to Pivot'
Garg believes app-based platforms such as Uber Technologies Inc. (UBER) and Airbnb Inc. (ABNB) will also need to adapt as AI agents become more capable of planning trips, booking transportation and making purchasing decisions without requiring users to manually navigate apps.
"Uber and Airbnb will also need to pivot," he said. "Without adapting to serve AI, as the number of human app users drops, they face losing out to competitors better prepared to serve the new agentic user base."
If there is a safer corner of the market, Garg points to the companies supplying the infrastructure powering the AI boom.
Why Nvidia, TSMC and Qualcomm Are the 'Safest Bet'
"TSMC, Nvidia, and Qualcomm are the safest bet," he said.
His reasoning is simple: regardless of which AI platform ultimately wins, agents still require computing power. That dynamic could benefit companies such as Taiwan Semiconductor Manufacturing Co. (TSM), Nvidia Corp. (NVDA) and Qualcomm Inc. (QCOM), which supply the chips and hardware underpinning AI workloads.
While Wall Street continues to debate which chatbot, model or platform will dominate the next phase of artificial intelligence, Garg argues investors should pay equal attention to a different question: which business models survive when AI agents start doing the clicking, searching and booking for us?
