How worried should Big Tech, investors be about Deepseek? UBS weighs in on AI's 'Sputnik Moment'

DeepSeek, a Chinese AI startup, has dropped a bombshell with its V3 and R1 large language models, promising top-tier performance at a fraction of the cost. The market’s reaction? Panic. NASDAQ futures tumbled 3% as investors scrambled to make sense of what this means for the titans of Silicon Valley.

According to a UBS report, DeepSeek’s cost-efficient approach has challenged Big Tech’s substantial capital investments. For example, Meta recently raised its 2025 capital expenditure forecast to $60–65 billion, $10 billion above consensus estimates, signaling its continued commitment to AI development. However, while DeepSeek’s models are impressive, they still lag behind leading-edge models from industry giants, and there is limited transparency regarding its training methods.

Read More

Marc Andreessen of Andreessen Horowitz compared DeepSeek’s breakthrough to "AI’s Sputnik moment," suggesting it could trigger a competitive race in AI innovation and pricing. However, UBS analysts advise caution, noting that the tech sector has weathered similar valuation resets over the past decade, often rebounding strongly within a year.

UBS analysts, however, urge investors to stay calm. They point to history as a guide: over the past decade, tech valuations have reset by at least 10% almost every year, only to rebound strongly within twelve months. The 2022 reset, driven by macroeconomic shocks, saw a 30% drop, but the sector roared back. The message is clear: volatility is an opportunity, not a threat.

As the industry braces for a wave of earnings reports—Microsoft, Meta, Apple, and others are set to report this week—all eyes are on how tech leaders will respond. Will they double down on innovation, or will DeepSeek’s disruption force a rethink of their strategies? One thing is certain: the AI game just got a lot more interesting.

“The real treasure of AI isn’t the UI or the model—they’ve become commodities,” says Marc Benioff, CEO of Salesforce. “The true value lies in data and metadata, the oxygen fueling AI’s potential.”

With $60 billion capex bets on the line and a history of tech rebounds, the stakes have never been higher. The question isn’t whether Big Tech will respond—it’s how.

What for markets?

• Negative market sentiment is emerging towards AI and Big Tech due to the success of China-based DeepSeek’s language models.
• DeepSeek’s cost-efficient model training has raised concerns about high capital expenditures by Big Tech companies.
• NASDAQ futures were down 3%, reflecting worries over the competitiveness of U.S. tech giants.

What about investors?

• Investors are advised to remain calm and await upcoming tech company earnings results for clarity.
• Volatility presents an opportunity for structured investment strategies, including buying quality stocks during market dips.
• Historically, corrections in quality tech stocks or benchmarks have provided good buying opportunities.

What about AI industry? 

• DeepSeek’s advancements could accelerate AI adoption and potentially bring artificial general intelligence (AGI) closer.
• Investors may need to reassess their AI portfolios, focusing more on AI applications and intelligence layers rather than enabling layers.
• Despite DeepSeek’s disruption, U.S. tech companies like Meta are increasing capital expenditures, signaling continued investment in AI.

Historical context

• Over the past decade, the tech sector has experienced at least one 10% valuation reset annually, with 2017 as the exception.
• Tech indices have typically rebounded strongly within 12 months of such resets, underscoring the sector’s growth potential.