Welcome to Stuff I Thought About Last Week, a collection of topics on tech, innovation, science, the digital economic transition, the finance industry, “AI”, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday EMAIL

In today’s post: ransomware attacks on hospitals are reaching urgent levels as Biden increases crypto scrutiny; Covid breathalyzers; products like Tesla that judge you; Lego and the loss of toy merchandising in a streaming world; TikTok explores shopping; Amazon's vulnerabilities; the major challenges to scale lab-made meat; AI-washing is reaching a peak, much like dotcom before it; and much more below.

Stuff about Innovation and Technology
Hospital IT Requires Intensive Care
A whopping 43% of healthcare organizations have been affected by a ransomware attack, with 22% reporting a subsequent increase in patient mortality (out of ~600 health providers surveyed), according to cybersecurity publisher SC Magazine. The cyberattacks also caused longer hospital stays and delayed procedures. The WSJ reported on a lawsuit over a potentially preventable infant death due to a hospital hack that caused a critical monitoring alarm to be missed. We are likely to see a Y2K-like IT upgrade frenzy to replace vulnerable legacy systems and modernize IT in the coming years. And, we may also continue to see governments banning and/or regulating crypto, a prime vector for ransomware payouts, to provide the time for these upgrades to happen. Regulation of crypto will anoint a small number of winners via regulatory capture. The Biden administration is convening a panel of 30 countries to determine how to crack down on ransomware attacks and reportedly intends to have stablecoin issuers regulated as banks. Covering the topic of ransomware attacks and security risks, NZS’ Joe Furmanski sat down with our friend Dex McLuskey to explain zero-trust security and how we think about investing in the sector relative to our Resilience and Optionality investment framework. We’ve been interested in the evolution of enterprise security to an identity-centric model ever since an investment in Okta in 2014. You can grab Joe’s podcast here or read a transcript here.

Covid Breath Test
Several companies are working on COVID breathalyzers to improve upon the current nasal swab testing for identifying infected individuals. IEEE reports on the technology, which could start popping up around the globe. Singapore has approved Breathonix – not to be confused with Breathomix, which has been deployed in the Netherlands. While I don’t fully understand the scope of the technology, I wonder if it could be adapted for monitoring in air handling systems as well?

Digital Dressing
Luxury fashion retailer Farfetch worked with digital fashion platform DressX to create 3D models of luxury fashion items, which can then be virtually fitted on influencer-provided photos. The practice is aimed at decreasing the physical samples sent out to fashion icons, gauging consumer interest before commercial production, and making the process of showcasing new items more efficient. And, of course, digitized items are just a click away from ordering.

Tesla Knows if You’ve Been Bad or Good
Before Tesla grants you access to the latest “full self-driving” release, you have to be deemed a good driver. As an interesting consequence of increased device connectivity and data collection, more companies will be able to dictate to what extent we can or cannot use their products. Imagine what kinds of consumer behaviors a company could influence.

Lego Up on the Competition
Lego has been crushing the competition, posting H1 2021 earnings of $1B – 10x more than the #2 Hasbro. Lego has executed brilliantly, innovating to keep up with its evolving customers. In decades past, toy lines were heavily entwined with kids’ linear TV shows, both for advertising and storytelling (see #234 for more details). As kids have migrated to YouTube and streaming, merchandising deals have lagged. With streaming platforms more interested in short-term hits than long-term franchises, and offering fewer ad opportunities, the days of building a toy line on the back of a TV show may be over. This shift favors brands like Lego, whose popularity doesn’t sink or swim with a TV franchise. If you haven’t seen the first two seasons of the reality competition show Lego Masters on Hulu (or wherever you watch your Fox TV shows), it's great family fun and remarkable to see what people can do with Lego.

TikTok on a Tear
TikTok has 1B monthly active global users and is closing in on Instagram, which probably has around 1.3 to 1.4B monthly active users (the last official disclosure from parent Facebook was 1B in 2018). TikTok did not disclose daily active users. In the summer of 2020, TikTok reported just under 700M global users, of which around 100M were in the US. Among Gen Z, TikTok is thought to have surpassed Instagram for monthly users. As noted in #304, TikTok garners more of US users’ time than YouTube and is the fastest growing social app. They also revealed more about their ad and shopping strategy at TikTok World last week.

Amazon’s AWS Margin is Your Opportunity
Many years ago, Jeff Bezos infamously said “your margin is my opportunity” as he was leading Amazon on a multi-decade growth path that challenged the fat margins of many incumbent competitors in retail and IT. Lately, when it comes to AWS, there are some folks that would like to turn that quote on its head, as Amazon has become the 800-pound gorilla in the cloud computing industry. The Information interviewed Ariel Kelman, who was AWS’ global marketing head for nine years and is now at Oracle. Kelman noted, in an ironic twist, that Oracle is now trying to sell themselves as the customer-friendly alternative to AWS and that AWS’ fees to remove data from the cloud platform are a “customer-hostile pricing strategy”. Cloudflare agrees. Last week, they introduced R2 storage on their edge computing platform with no fees for data removal. A series of blog posts from the company calls out Amazon’s massive markups (reportedly 80x for US/Europe) and their “egregious egress” pricing. Obviously, this is all just a marketing war, but the idea that Amazon would have left the door open to disruption by becoming the high-cost option would have been unthinkable a decade ago. In IT, almost nothing ever dies (just look at IBM mainframes!), but growth rates can stagnate if developers change platforms. Workloads are changing to serverless functions, and proximity to users is becoming more important for low latency. Today’s cloud platforms can continue to be successful because typical workloads are still growing very rapidly, but they aren’t necessarily architected toward these newer types of apps. If you want to know where the market is headed, always follow the developers.

Amazon’s Hardware
Amazon has been focused on consumer hardware development both internally and through acquisition. The obvious successes are the Alexa smart speaker/screen devices along with the acquired Ring security and eero WiFi product families. The more cutting-edge hardware, however, has seemed more vaporware than tangible. Last year, they showed off the Ring in-home security drone (which I longed for in #264), but they are only now saying it’s open to private testing. According to Ring founder Jamie Siminoff: “I have it in my home, and it does work, but today’s homes are so unique, so we really need to get it into more customers’ homes to make sure everything we are doing is right. With any other product, we would have probably just been shipping. With this one, we are going to take our time, make sure it’s right before we go to full, general availability.” In what seems like another potentially too-early-for-primetime announcement, Amazon unveiled a little Alexa on wheels called Astro. Here’s a description from CNBC: “Astro is about the size of a small dog. It roams around your house on three wheels, including two big ones that prevent it from getting stuck and a smaller one for rotating. It has a camera that rises up on a 42-inch arm that can keep an eye on your home as Astro patrols while you're away. It can follow you around and play music or display TV shows on its 10-inch touchscreen. It can recognize faces (if you want it to) so you can load up two sodas in the back storage compartment and tell Astro to go to someone in the living room.” I like the imagination, but I'd like it even better if the products were fully baked.

Amazon’s Big B2B Business
Rounding out Amazon news, Industrial Distribution reports that Amazon could be the largest industrial distributor, with $27B in projected B2B marketplace sales this year. Because Amazon doesn’t disclose Amazon Business sales or break them down by end market, some of this total might just be landlords ordering light bulbs, so Grainger officially stays at the top of the list at less than half of the $27B in sales. Some distributors have grown alongside Amazon by focusing on proximity to customers for rapid fulfillment (such as Fastenal’s onsite parts vending machines), but it does seem likely that Amazon has siphoned off some share of distributor growth. Amazon is also pitching a broader procurement and cloud solution for large enterprises vs. what most product distributors are offering.

Miscellaneous Stuff
“Mosquitogeddon”
Climate change is helping mosquitos thrive and spread disease. A temperature rise of only 1 °F increases humidity by 4%, helping to supercharge storms. The resulting record-breaking rainfall has significantly expanded habitat and breeding opportunities for the pesky bugs. If you need one more reason to become a climate activist, let this be it!

Cultured Meat Pie in the Sky
Lab-grown meat still faces huge real-world scaling problems, as detailed by The Counter and Biotechnology and Bioengineering. To start, a massive, hypothetical bioreactor facility, equivalent to ~1/3 of all of the volume of the biopharmaceutical industry today, would only yield 22M pounds of lab-grown protein, or around .02% of the 100B pounds of meat the US produces each year. To ramp cultured meat production to 10% of global demand projected for 2030, we would need 4,000 factories of that size encompassing 2.4M bioreactors at a cost of $1.8T. And, even at that scale, the cost per pound might end up multiples higher than conventional animal husbandry. One of the problems with animal cell culture is the risk of viral and bacterial infection, requiring laboratory-grade cleanrooms. When you are talking about supersized bioreactor arrays, the costs for building the cleanrooms alone would be extraordinary. And, cleanrooms only reduce, but do not eliminate, the chance of infection. Since cultured cells have no immune system and are relatively slow growing (compared to viruses and bacteria), a single point of failure would cascade, causing a potentially catastrophic loss in material and downtime. And, even with pristine conditions, it’s not at all clear that animal cells would even be amenable to that kind of scaleup production, which has never been attempted. They don’t like to grow too densely, require ongoing supply of fresh nutrients and waste removal (lest their catabolites, like ammonia and lactate, build up and inhibit their own growth), and are fragile to shearing if overly agitated. Also, when you look at the costs (and required amounts) of media ingredients, such as amino acids, micronutrients, etc., it starts to look like animal-derived protein isn’t so inefficient after all. The cultured meat conundrum continues to garner a lot of attention globally, and, eventually, we may see breakthroughs with the media ingredient supply chains, engineering of more robust and densely-growing cells, and greater bioreactor size/efficiency. Not every pie-in-the-sky dream makes it past the engineering hurdles, but history is also littered with pessimists who didn’t see the viable path for disruptive innovation that can be forged when humans put their minds to something. For now, however, it seems skepticism is still in order with respect to lab-grown meat.

Stuff about Geopolitics, Economics, and the Finance Industry
AI is the New Dotcom, and That’s OK
Back in the late 1990s, every business was either appending “.com” to their existing name or touting their dotcom strategy and how it was going to transform them or their industry. A lot of hyped-up ideas ended up being right, just twenty years too early. But, for the many legacy companies that put on dotcom lipstick at the turn of the century, the Internet was ultimately a negative disruption of their business. For some industries, such as media and retail, we’ve seen the near completion of the disruptive, Internet-enabled transformation. For more highly regulated businesses, such as the banking and healthcare sectors, which have successfully lobbied to keep disruption at bay, it’s unknown if/how they will ultimately be affected by the Internet Age. And, for a large bucket of companies that have harnessed the Internet to improve their products, supply chain, and/or customer interactions without significant disruption to their business model, dotcomization has been more subtle. For all industries, the Internet enabled an accelerated pace of change, and dotcom simply became shorthand for digital transformation. The biggest winners of the Information Age have been the new companies – those that were built by the Internet, for the Internet, in the late 1990s and early 2000s. These are the familiar mega platforms of today, all of which are facing a global wave of regulatory pushback that is handicapping their ability to enter new markets or even innovate in some of their existing markets. These constraints are likely to lead to paralysis for many companies, as was the case for Microsoft nearly two decades ago (they were able to recover, but it took a long time).

And, now, we find ourselves in another technological jargon bubble: AI. For most established companies today, AI of course is no more real than dotcom was twenty years ago for Industrial Age companies. AI has jumped the shark and become a metaphor for digital transformation in exactly the same way dotcom evolved. For most of today’s companies, AI will be a negative disruption – an acceleration of the digital transition that will bury the incumbents and create new winners. And, AI, just like dotcom before it, will be enabled by software and semiconductors, the enduring engines of the analog-to-digital transformation of the global economy. It’s possible that the emergent dotcom winners, whose cloud platforms constitute the next semi/soft incarnation, will power the new AI wave of innovation and transformation. And, there will be new companies, just like twenty years ago, that will dwarf anything we’ve seen in the last two decades. There remains a chance today’s mega platforms will maintain power, but that’s up in the air with regulation and the challenges any incumbent faces when a new market emerges. AI also has a mega investment bubble associated with it. So far, this bubble has stayed more in private markets, while the dotcom bubble manifested a little faster in the public markets of the late 1990s.

There is one big difference I see with AI as the new marketing term for digital transformation compared to dotcom: we seem to be willing to believe AI is real, but it’s not. At least not yet. We are seeing companies hand off decision making to AI as if it were something more elevated than a complicated search algorithm scanning pools of data. It’s this dangerous “invisible hand” mentality of AI – which Jaron Lanier warned us about – that has me worried. It’s distorting how people are hired and fired, whether people can rent apartments and access government benefits, and how doctors and hospitals treat patients. It’s determining who people date, what music they listen to, what shows they watch, and who they interact with. Facebook was designed to collect, amplify, and reflect our own thoughts and emotions back to us in order to sell ads. In and of itself, social media is a dangerous reflective chamber, but becomes even more so if we accord its half-baked algorithms any sort of divine power. All of these life changing algorithms have the transparency of mud, and to rage against them is to shout into an infinite abyss. AI is nothing more than a fallible human tool – it’s a reflection of human nature, and lately, we’ve been a little crazy. Along with the increased influence of social media algorithms, we’ve seen a rise in extremism, hate, misinformation, and violence. There will of course be legitimate artificial intelligence systems and uses that perform incredibly complex tasks and might be able to make short-term predictions and decisions that are useful. But, recently, when I see a company talking about AI, that is clearly not what they mean.

We will have incredible breakthroughs in artificial general intelligence when AI can gather enough context to make analogies, reason, and learn on its own. However, the computational burden of trying to make better decisions through deep learning systems might also hit a breaking point even with these breakthroughs. IEEE reports that halving the error rate in image recognition would require 500x the computational power! The world is complex and unpredictable by nature. The amount of energy required just to predict the weather an hour from now, on every square meter of the planet, might just not be worth it compared to looking out the window. I can’t help but laugh at the absurdity of DeepMind’s new rain prediction nowcasting model, which trained on 16 TPU cores for one week, accelerating global warming in exchange for a mere 1-2 hour advance notice of impending storms (fairly useless for staging a massive evacuation). I wonder whether their model takes their own climate impact into account! The butterfly effect is real.

The next wave of digital will bring major structural changes to the economy as well with deflationary pressures and job displacements. There will be periods where the short term negatives outweigh the long term positives. Long-term, I am optimistic about the ongoing, accelerated, digital transformation of the global economy. It will ultimately do a lot of good for people and the planet over the next few decades, but I am a little worried about the blind faith that’s being put into what amounts to nothing more than a marketing term. So, if you are a CEO, or marketing person, appending “AI” to your company, product, service, etc., I still want you to do that – I love the benefits of the accelerated use of technology – but please recognize you are dotcomming your business. Just like twenty years ago, there is a lot of work to be done; but, decades from now, we will be looking at a completely digital, completely transformed economy, perhaps with a little bit of real AI.

✌

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.