MBI Daily Dose (July 07, 2025)

Companies or topics mentioned in today's Daily Dose: Fiscal tightrope, Productivity boom, ChatGPT traffic, History of Intelligence

UBS published this interesting table comparing and contrasting US’s current high debt level to similar situations in the past. Considering the recent US administrations’ poor fiscal discipline, an AI-driven surge in productivity may now be the one of the very few (only?) “get-out-of-jail-free” card left.

Speaking of AI led productivity boom, we can be quite encouraged to see the recent productivity numbers in the US which is noticeably above pre-2020 trend. Any comparison about pre and post-2020 is, of course, quite challenging given how many key variables decidedly shifted at the same time. This blog post mentions three such variables that likely contributed to productivity boom seen in numbers: rise of work from home, higher job switching to match your skill, and rise of entrepreneurship post-2020. Despite these factors, it is AI which is likely leading the productivity boom:

…let’s remember if this surge were caused by work from home, labor reallocation, or more startups, we might expect to see broad-based, one-time-boost-type gains across many industries or concentrated in sectors with more work from home, et cetera. But that’s just not really what drove it. If you look at the industries experiencing the most significant productivity surge, seven or eight out of the top ten look tech or AI intensive … We’re talking about things like internet publishing, e-commerce, computer system design, renting intangible assets, motion picture and sound recording, and miscellaneous professional, scientific, and technical services.

ChatGPT’s traffic to news sites is exploding while traffic from traditional search is declining. It’s also interesting to note how stocks and finance related topics dominate the traffic. You may no longer need to make up an explanation on your own to your PM’s query “why is this stock up/down 3%?”. From TechCrunch:

From January through May 2024, ChatGPT referrals to news sites were just under 1 million, Similarweb says, but have grown to more than 25 million in 2025 — a 25x increase.

Topics like stocks, finance, and sports are currently accounting for the majority of these ChatGPT news-related prompts, but Similarweb’s report notes other topics are seeing growth, too, like politics, the economy, weather, and others.

This, the firm theorizes, may signal a move away from more “reactive information” and toward deeper “issue-driven engagement” via AI.

In addition to "Daily Dose" like this, MBI Deep Dives publishes one Deep Dive on a publicly listed company every month. You can find all the 60 Deep Dives here. I would greatly appreciate if you share MBI content with anyone who might find it useful!

While driving around Big Sur yesterday, I was listening to this podcast which interviewed the author of the book: “The History of Intelligence”. I didn’t need a coffee break thanks to this podcast as my jaw dropped several times while listening to it. I recommend listening to the podcast, but I asked Gemini (I’m usually a ChatGPT user but noticed Gemini is better if I ask it any query on YouTube videos) to summarize some of the studies mentioned in the podcast. I’m copy-pasting Gemini’s response to a couple of the studies below:

The Devaluation Study on Habit Formation

This classic experiment is used to distinguish between goal-directed actions and ingrained habits. A mouse is first trained to press a lever to receive a specific food treat. Later, in a separate environment, the mouse is given that same treat but laced with a substance that causes sickness, devaluing the reward. When the mouse is returned to the original box, its behavior reveals its motivation. If the training was brief, the mouse will avoid the lever, showing it connects the action to the now-undesirable outcome. However, if the mouse had performed the action hundreds of times, it will often continue to press the lever compulsively, demonstrating the action has become a rigid habit that is independent of the goal it once served.

Emil Menzel's Chimpanzee Deception Study

Researcher Emil Menzel was initially studying spatial memory in chimpanzees. He would hide food in a large, one-acre enclosure and show its location to only one chimp, a subordinate female named Bella. At first, Bella would lead the group to the food, but a dominant male named Rock began to aggressively hoard the treat. In response, Bella's strategy evolved. She first began waiting for Rock to be distracted before retrieving the food. Eventually, she started to actively practice deception, deliberately leading the group in the wrong direction to mislead Rock and secure the treat for herself, demonstrating a sophisticated capacity for strategic thought and planning.

Current Portfolio:

Please note that these are NOT my recommendation to buy/sell these securities, but just disclosure from my end so that you can assess potential biases that I may have because of my own personal portfolio holdings. Always consider my write-up my personal investing journal and never forget my objectives, risk tolerance, and constraints may have no resemblance to yours.