Tracking DeepSeek on Earnings Calls: From Zero to 100+ Mentions

Hey, it's Matt. Welcome to AI Street Markets, where I break down AI investing tools. Was this email forwarded to you? Sign up below:

NEWS

I try to do two things in this markets newsletter

1. Explore new AI tools for investment research

2. Tie that analysis to market trends in the news

If you’re new here, check out the last three editions for background/ tips:

• Demo: AI Agents in Market Analysis

• ChatGPT's Task Scheduling: Automated Market Analysis

• Effective Prompts For Investment Research

*As a reminder, none of this should be construed as investment advice. I’m just showing some compelling ways that you can use AI for research.

DEEPSEEK TIMELINE

This week, AI startup DeepSeek rattled markets with its open‐source chatbot that basically matched the performance of proprietary models like OpenAI’s at a fraction of the cost.

Shares of Nvidia plunged nearly 17% on Monday Jan. 27, wiping out almost $600 billion in market value—the largest one-day loss ever recorded.

But markets tumbled a week after DeepSeek unveiled its new R1 model, which came out on Jan. 20.

The AI community grasped its impact faster than investors. On January 21, a day after DeepSeek released its new model, LLMWatch published an article titled DeepSeek-R1: What It Is & Why Everyone Is Talking About.

A few days later, a blog post from a former hedge fund equity analyst titled “The Short Case for Nvidia Stock” caught the attention of investors with a detailed, thorough, and accessible analysis of what DeepSeek means for the world of AI.

For this issue, I’m going to explore whether DeepSeek was on the radar of any company by using Aiera, an AI platform for investors that analyzes earnings calls and events across 13,000+ companies.

Here’s how I searched:

And you can see that it came out of nowhere:

DeepSeek started gaining traction after its mention at the World Economic Forum on Jan. 22. From there, it quickly became a hot topic, earning over 100 mentions in earnings calls and events tracked by Aiera.

To dig deeper into DeepSeek’s rapid rise, I used Aiera’s data with Claude, an OpenAI competitor from Anthropic, to highlight key executive quotes.

Here’s how some of the biggest names in AI and investing reacted:

• Satya Nadella, CEO of Microsoft: "DeepSeek has had some real innovations... now that all gets commoditized, and it's going to get broadly used. And the big beneficiaries of any software cycle like that is the customers."

• Mark Zuckerberg, CEO of Meta: "I think there's a number of novel things that they did that I think we're still digesting. And there are a number of things that they have advances that we will hope to implement in our systems."

• Arvind Krishna, CEO of IBM: "We have been very vocal for about a year that smaller models and more reasonable training times are going to be essential for enterprise deployment of large language models... As other people begin to follow that route, we think that this is incredibly good for our enterprise clients."

• Jonathan Gray, President of Blackstone: "We've obviously been spending a lot of time the last week looking at the impact of DeepSeek. I'd start with our data center business, which is the largest in the world. We have $80 billion of leased data centers."

I also asked Claude to use the DeepSeek mentions to answer the following questions in the below prompt:

1. How did different sectors/industries react to DeepSeek's announcement?

2. How did companies with significant AI investments frame this development in relation to their existing strategies?

3. What were the main perceived implications for AI infrastructure and data center demand?

This prompt generated the following analysis:

I'll analyze the reactions and implications from the earnings call transcripts:

1. Different sector/industry reactions:

• Semiconductor/Hardware Companies:

  • Samsung (005935) focused on HBM (High Bandwidth Memory) implications, taking a measured stance on impacts to memory demand

  • Intel (INTC) saw it as an opportunity to leverage their diverse portfolio of CPUs, GPUs, ASICs, and FPGAs

  • ASML was questioned about impacts on advanced chip demand

• Infrastructure/Data Center:

  • Blackstone (BX) emphasized their $80B leased data center portfolio remains secure with long-term leases

  • Flex (FLEX) maintained bullish outlook on data center power infrastructure demand

  • Brookfield Infrastructure (BIP) faced questions about data center development strategies

• Software/Platform Companies:

  • Microsoft (MSFT) framed it as validation of their software optimization work with OpenAI

  • ServiceNow (NOW) viewed it positively as model commoditization could benefit application layer companies

  • Meta/Facebook (META) saw it as validation of their open-source AI strategy with Llama

2. Companies with AI investments framed it as:

• IBM (IBM) called it a "point of validation" of their strategy focusing on smaller, more efficient models

• Meta emphasized how it validates their open-source approach with Llama

• Microsoft highlighted how it aligns with their work on inference optimization

• ServiceNow positioned it as beneficial since competitive differentiation will happen at the application layer

• Coursera (COUR) saw it as positive for democratizing AI development

3. Main implications for AI infrastructure:

• Data Center Demand: Most companies (like Flex, Brookfield) maintained bullish outlook on data center growth

• ABB reported data center business grew from 12% to 15% of order intake and expects continued growth

• Blackstone emphasized their data centers are long-term leased and they don't build speculatively

• Several companies indicated this could accelerate AI adoption by lowering costs

• Companies like Flex noted it could drive more distributed computing needs

• Power/cooling infrastructure demands expected to remain high despite efficiency gains

• General consensus is that while individual models might become more efficient, overall infrastructure demand would grow due to broader AI adoption

The prevailing view across sectors was that while DeepSeek might change how AI is deployed, it would likely increase overall AI adoption and infrastructure needs rather than reduce them.

Everyone is scrambling to interpret what DeepSeek’s emergence means, but the reality is, its long-term effects are hard to predict. Companies with major AI investments are naturally framing this as a net positive and it may well be.

One thing seems clear: five years from now, AI will be even more embedded in our lives.

PODCAST

Alpha Intelligence Podcast: Unpacking DeepSeek

I just launched the Alpha Intelligence Podcast with my co-host, Francesco Fabozzi, featuring in-depth conversations with hedge fund managers, quants, and academics on the future of AI on Wall Street.

For the first episode, we spoke with Tharsis Souza, PhD, former Senior VP of Product Management at Two Sigma, Lecturer at Columbia University, and author of an upcoming O'Reilly book on LLMs' pitfalls.

Our discussion focuses on DeepSeek’s latest model, R1. While DeepSeek claims to have trained R1 for just $6 million without Nvidia’s most advanced chips, the reality is more complex.

What you'll learn:

• On the podcast, we break down why DeepSeek’s $5.5M training cost might not tell the full story

• We also talk about the big shift in AI compute and what it means for companies betting on infrastructure.

• Learn why "LLMs are the trappiest tool ever for finance people" and how to think about look-ahead bias when using AI in financial applications

• Hear why "the price of intelligence is going to zero" and how to identify opportunities that will become possible as models get stronger

Listen on Spotify or YouTube