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

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: grieving with the simulated dead; changing accents in real time speech; automation has been painfully slow to come to ecommerce; should we want companies to have opinions and incentivize their outcomes?; is TikTok displacing music streaming?; quantum batteries; VC-backed gig economy jobs may be at risk, spelling relief for hiring in the service economy; and, much more below...

Stuff about Innovation and Technology
Sim Relatives
Amazon’s Alexa voice assistant division announced it is working on a technology that will mimic anyone’s voice after processing only about one minute of audio from that person. One of the goals is to let people hear loved ones who have passed away. In the demo, Alexa reads a book to a child in the voice of his grandmother. It’s not a stretch to think that technology already available today would allow us to have actual conversations with the dead based on everything that person said on social media, emails, videos, and other sources of information on their life and views. And, from there, it’s not impossible to imagine this technology being loaded onto a lifelike robot. If this scenario sounds familiar, it’s because it was the plot of “Be Right Back”, the first episode of the second season of Black Mirror. After the Alexa announcement, several articles noted the dangers of criminals using the feature to copy voices, but the technology to do so is already widely available. I am more interested in whether it would be easier or harder to grieve the loss of human life if you could continue to carry on seemingly real conversations with the deceased? Will we need to give consent to our descendants in order to use our voice from beyond the grave? At the same event, Reuters reports, Amazon divulged that they had around 100M Alexa users – a notable lack of any growth since 2019, which may highlight the relatively small island of Amazon customers compared to the massive billion+ user bases of the larger Android and iOS platforms. In the mega race to create a working chatbot – which I think will ultimately be the new platform interface for everything we do (including AR companions, see #332) – gathering data from people speaking to simulated relatives would be quite valuable.

Morphing Accents
In related news, a startup has developed technology to change accents in real time while preserving the underlying voice. The software from Sanas, already in use in some call centers, can transform any accent into any other type of accent. Here too we can see that the data collected from a massive number of conversations could be quite valuable to the development of chatbots and the next computing UI.

Replicating Humans is Really Hard
Amazon reportedly was worried about running out of workers by 2024 based on the growth path they identified during the pandemic. Of course, we now know that projection was largely based on an error in their forecasting systems, an example of the pervasive recency bias we saw in the pandemic. At some point though, as the analog-to-digital transition of the economy creates natural power laws, Amazon might again find itself facing a worker shortage. This threat puts the burden of filling the labor gap on automation. On this front, Amazon has always struck me as being well behind the curve. While they excel at software automation, they have largely built their logistics around people. Even with warehouse robotics, they’ve only just announced that they have an autonomous pallet-moving robot that can work alongside people (previously, Amazon fulfillment center robots only worked in caged-off areas). It’s been over ten years since Amazon acquired the warehouse robotics startup Kiva with the goal of modernizing logistics with such free-range, autonomous, “human equivalent” bots. The pace of progress seems painfully slow. Part of the issue is the complexity and fragmentation of the robotics industry and use cases, as well as the lack of a platform. And, of course, humans are still just more efficient at most tasks. McDonalds can’t even get their speech recognition system to recognize a simple set of menu items for automated ordering (this automation failure, however, might speak more to McDonald’s inability to properly leverage technology). I’m not holding my breath for the Tesla Optimus prototype, said to be coming this September for the company’s AI Day, but it’s perhaps such lofty ambition that will bring us closer to a general-purpose human-labor replacement (let’s just hope Optimus doesn’t meet Alexa and try to recreate "Be Right Back"!).

Corporate-Backed Behavioral Incentives
Back in the fall of 2019, I wrote about the potential for software companies – given their pervasive importance to the business world – to incentivize their desired social outcomes (#210). Companies incentivizing customers with discounts could be more impactful than investors trying to convince boards to improve their ESG scores. What if, for example, Microsoft gave a ten percent discount on Office 365 for customers that pledge to achieve carbon neutrality at some future date? A fairly benign example (which I don’t think has had much of an impact) is Starbucks’ 10c discount (and 25 star points) for bringing in your own reusable cup (which is completely thwarted when you digitally order ahead!). There are certainly companies that have held public opinions on issues from their very start. Chipotle, for example, has views on animal welfare that not everyone might agree with, but, if you do, you can choose Chipotle over Taco Bell (unless you want a Taco Bell Mexican Pizza, which is always forgivable). There’s a slippery slope for incentives and I also understand the arguments for corporations avoiding political and moral decrees altogether. Let’s definitely not leave moral judgements to CEOs! It should be acceptable for companies to remain apolitical and not face cancelation if they don’t take a stance. The free market seems to generally (eventually) land in the right place, and in some ways that's what I am talking about here: would companies see positive or negative impacts to their businesses if they incentivized behavioral changes for their customers? In our increasingly polarized society, the safest path is for companies to stay quiet, but our current lack of inspiring leaders across the public and private sectors has left a vacuum. Lately we've seen a lot of CEOs make grand statements only to do nothing after they get their soundbites in the press. Societal division seems to remain the only path for now, and maybe that's not a problem corporations could, or should, solve.

Short-Form’s Long Arm
Subscriptions to music services like Spotify declined 1M in the UK in Q1 this year, with subscribed customers under 35 falling from 57% to 53.5% y/y, according to Reuters. While 37% of survey respondents indicated economic weakness was the primary reason, I wonder how much the rise of short-form video is impacting media habits – eating into music listening as well as long-form video viewing. Last week, I lamented the shift in human storytelling from long- to short-form. Regardless of what’s behind the decline, the drop in music subscriptions surprises me. One way long form video can reclaim people's attention is to produce a smaller number of much higher quality shows that are capable of holding our attention (like the new series The Old Man with Jeff Bridges and John Lithgow on Hulu and FX). Ultimately quality should win in art, but right now addictive apps dominate. Meanwhile, Bloomberg reports that music labels are increasingly engaging with TikTok, and TikTok itself has interest in signing artists and becoming a label.

Miscellaneous Stuff

Quantum Batteries
Researchers are just beginning to investigate the real-world potential for quantum batteries. The still-theoretical devices use excitable (e.g., photosensitive) molecules to store energy. Because quantum systems are probabilistic rather than deterministic, each molecule’s transition amplitude – the probability of going from a ground to an excited state – can constructively interfere with the rest, giving a total amplitude that’s greater than the sum of the individual amplitudes. This type of “collectivism” doesn’t exist in a classical battery, where the individual cells function independently. Owing to this unit function, a quantum battery can undergo faster charging (because the cells can charge collectively rather than individually), as well as superextensive charging, which means the more molecules in the system, the shorter the charging time. Taking advantage of this collective action requires either quantum entanglement (which is quite fragile) or quantum coherence (which is more stable and doesn’t require temps near absolute zero). Additionally, since coherence is easier to control, it allows for selective triggering of decoherence to modulate the rate of battery discharge. Coherence can also preserve the integrity of the charged state by preventing energy leakage into the environment. The practical implementations here are many years into the future, but, until then, at least you don’t need to worry whether your EV is existing in a simultaneously charged and uncharged quantum state.

Stuff about Geopolitics, Economics, and the Finance Industry
Will Rates Curb Gig Economy?
The FT wrote an article about the loss of startup funding titled: “Farewell to the Servant Economy”, which got me thinking about labor shifting away from the VC-backed gig economy. As interest rates rise and the most speculative growth business models risk losing funding, there’s some question as to how big of an impact this might have on employment. According to a December 2021 Pew study, 16% of Americans reported earning money from online gig platforms. While perhaps not all of these jobs are at risk, certainly a very high percentage of them are tied to startups that are not profitable. As I wrote about pre-pandemic in #215, a large degree of modern life was becoming subsidized by VC via companies that were aspiring to get to scale in order to stop losing money. Few of those companies made the leap to scale, and even fewer, if any, have become sustainably. The fundamental issue is the cost-of-labor transfer. As I noted with respect to food delivery in #309: “It may simply not be possible to transfer the labor cost from the shopper/eater to the merchant given that food is such a low-margin, highly-fragmented industry overall.” Grocery shopping, including drive time, can be easily 45 minutes to 1 hour, or $10-20 in labor costs. Only a small subset of the population can transfer that much of their own labor to someone else and pay enough (the average grocery store tab is $40-50, so that’s a 20%+ cost increase) for the company facilitating the labor transfer to make a margin. Ultimately, such services may be just for the rich until automation and technology can dramatically reduce costs. But, back to the employment problem, Pew reported that 30% of 18- to 29-year-olds have worked for a gig-economy company. While that includes a range of part-time to full-time earners, it seems plausible that several percent of the workforce would be at risk of losing part- and full-time jobs as startups run out of funding with rising rates. If this cohort of workers returns to traditional service jobs (with salaries paid by companies that earn profits), it could go a long way toward resolving labor shortages for the industries competing for workers with on-demand companies.

✌️-Brad

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.