Brandertainment, Socialization at Scale, & Findings of Interest

This edition explores implications of a new variant of entertainment emerging, and one of the AI threats that matters most because we can't even feel it happening.

Edition #21 of Implications.

• This edition explores: (1) the peculiar emergence of “brandertainment” as brands compete and partner with traditional Hollywood, (2) the AI threat of socialization worthy of concerns, and (3) some interesting findings and surprises at the end, as always.

• If you’re new, here’s the rundown on what to expect. This ~monthly analysis is written for founders + investors I work with, colleagues, and a select group of subscribers. I aim for quality, density, and provocation vs. frequency and trendiness. We don’t cover news; we explore the implications of what’s happening. My goal is to ignite discussion, socialize edges, and help all of us connect dots.

• If you missed the big annual analysis or more recent editions of Implications, check out recent analysis and archives here. A few highlights include what to make of models displacing models, the great personalization wave, and chronicling a new phenomenon of small teams that build big businesses. Let’s go.

Brandertainment: Brands will create more mainstream media in direct competition with studios.

• There is much discussion about AI disrupting Hollywood and the rise of Indie moviemakers as the cost structure of production goes down, but less discussion about the pivots brands will make in a world where we become increasingly numb to meaningless marketing content and allocate our attention to things that entertain and inspire us. While brands have always wanted to engage us, the world of “Hollywood” was very much an old school network and constrained by the massive expenses associated with creating entertainment. But now, in an era with new AI capabilities, democratized idea generation, and a plethora of new distribution options for entertainment (the movie theater distributors are no longer the gatekeepers), the window has opened for new production sources of high quality entertainment. My bet is that brands will seize the moment to differentiate themselves and engage loyal fans through the medium of entertainment, possibly at the expense of traditional advertising.

• Over the last couple of years we have seen Mattel leverage their age-old Barbie IP with the launch of a new billion dollar media franchise for Barbie starting with their hit movie. We’ve seen the NFL start producing mainstream content including “QB” and the “Air” movie chronicling the rise of Nike. Crayola recently launched a series of efforts to develop original content for families. But I think we’re just at the beginning, and we’ll also start to see entertainment and brand marketing mutate into one another as brands use their IP and outsource script development, leverage custom AI models to design new characters as well as other modern tools that drastically reduce the cost of production (thus increasing their tolerance for risk). Fast forward 5-10 years, I anticipate a material amount of popular movies, streaming series, and podcasts will be produced by brands leveraging (and simultaneously promoting) IP from their primary business.

• As this happens, the economic calculus for producing entertainment will change as brands gain exposure for their IP (and, in some cases, conversion while watching when it comes to fashion, toys, etc) on top of box office sales and revenues from streaming rights. Further fueling this trend is the great wave of personalization coming to digital experiences of all kinds, including entertainment. When the articles of clothing worn by actors can be switched up using AI in post-production to accommodate different cultures and segments of potential customers viewers, brands will get even more ROI from producing high quality professional entertainment. Next, as mixed reality goes mainstream, entertainment will start to extend beyond the show or movie into our living rooms, social media, and shopping carts. All the more reason for consumer brands to consider entertainment as a new extension.

The AI threat we should be talking about: Scaled socialization as a service.

• Upon the recent attack against Donald Trump, the stark differences in the algorithms behind X and Threads were on full display. While these social media products are quite similar when it comes to basic interface and user experience, they are becoming increasingly different through the algorithmic choices being made by the operating teams behind them. The X platform seems optimized for what i’ve come to call “sensation engagement,” which includes politically polarizing content that catches any belief or sensitivity we each have by the edge and brings us down rabbit holes we just can’t bare to avoid. In contrast, the Threads platform seems optimized around existing and likely social connections and general interests as determined by past engagement. While I am only an “n = 1,” upon a major “sensational” polarizing moment, i’ve seen these two platforms portray wildly different dimensions of the world we live in, despite having similar social graphs on both. Behind the scenes, I suspect very different product decisions are being made. Perhaps X is trying to drive engagement by placing the spotlight on the loudest and most polarizing voices, while Threads is trying to be the source of rational diversified content more akin to a social network like Twitter of the old days?

• But the key realization here is the fact that algorithmic tuning has become the new social media product playbook for differentiation in the AI era. The fact that this is possible and so powerful reminds us that social media has the ability to change minds at scale. While we’ve been concerned about TikTok’s potential for influencing young minds for quite some time, I am struck by the rapid maturity of the “knobs” used to tune these algorithms. My greatest fear is that AI enables extremely effective socialization at scale that threatens the mere existence of democracy. Much like LLMs are managed by prompts, the proverbial “knob turners” running these social platforms have a conceptual “text-to-make-people-believe-something” prompt capability. To make matters worse, our feeds are powered by AI with increasingly long context windows that remember everything and everyone we engage with — and use this information to further optimize the intended outcomes of the knob turners.

• While I am a bonafide “accelerationist” in most respects (and generally believe that humanity adjusts to its tools in ways that ultimately benefit humanity), is it reasonable to fear how AI will enable extremely effective socialization at scale? I am not concerned about robots conquering the Earth, but I am gravely concerned about AI being used by humans to socialize the masses about dangerous mistruths, and rampant polarization to serve specific political agendas. The maturity of the “knobs” to tune these algorithms to engage and influence us in very specific ways really blows me away, as we discussed previously in edition #12. Especially as context windows (the memory that AI has of our questions and interactions) grow, we will have no idea when and how we are being socialized towards a particular belief by the seemingly endless and random feeds of our lives. Fast forward a few years (or a few weeks/months), are we liable to become lemmings of sorts, following the lead of whoever controls the knobs? Our industry has a history of being very creative about what can go right, and not as creative about what can go wrong. So, this provocation is merited.

• One possible implication: Socialization becomes a nefarious example of AI abuse. While political agendas have always existed, the means to spread them and subconsciously socialize people at scale has never been possible. But I believe we have reached a turning point given the effectiveness of algorithms. At best, we will all be further engaged to the point where ads become more effective and more valuable. But at worst, we will become socialized to extreme views at scale in ways we will never be able to understand or diagnose on our own. This “mass socialization” is my greatest fear in the age of AI. We don’t discuss what symptoms of “effective socialization” look like among US-based platforms and whether/how this power should, dare I say, be regulated.

• A more optimistic implication: A variety of consumer-oriented AI tools and native OS-capabilities will likely emerge that protect us from these negative implications of AI. For instance, imagine next generation browsers like Arc from The Browser Company or modern mobile operating systems that have their own AI and other forms of pattern recognition to surface our own vulnerabilities and actively merchandise the risks to us. Imagine your mobile device telling you when certain behaviors seem to be overly influenced by content you consumed. Imagine a browser disclosing, on behalf of a website, what AI and algorithms are being used either through disclosed or crowd-sourced information. Imagine AI Agents that always show the other side of a story, prompt questions we should ask ourselves before making a comment or clicking on a link, or tell us when we need a breather based on our behaviors using the device.

• One final and inevitable implication: The discipline of product, especially for consumers, will increasingly be about the knobs and how they are turned. Fine-tuning AI and developing the knobs that control user behavior is a new genre of product that we need to keep an eye on.

Assortment Of Findings

• Long-bets: I recently discovered this self-described “arena of accountable predictions” that I find absolutely fascinating. This platform allows people to make predictions and actually agree or disagree through the form of bets. Perhaps you’re betting that, “Within ten years - at least a significant minority of physicists will conclude that special relativity theory, as it is presently understood, has to be amended.” Or that “the S&P 500 will outperform Bitcoin by the end of 2140 (the end of Bitcoin mining).” Well, you’ve found a forum designed for long-term credibility as well as constructive brainstorming and debate among people that share a fascination for what the future holds.

• The 5 Types Of Wealth: My friend Sahil has spent the last few years writing a book all about rejecting the default ways we live life, especially around those that may involve money. In this new project, he poses frameworks and compelling arguments about how we should navigate what he calls “the five types of wealth,” namely Time Wealth, Social Wealth, Mental Wealth, Physical Wealth, and Financial Wealth. Some of Sahil’s ideas have had an impact on my perspective over the years - and I’ve also seen him jot insights down in his notebook for years - so I know this book is chock full of gems. It is available now for pre-order.

• Breaking the frame of “realistic” timelines: If you’re not signed up for the free ActionDigest, I highly recommend this one. One recent edition shared the story of Sylvester Stallone in 1975 sitting down to write a screenplay and, three and a half days later, emerging with the full script for Rocky. “At that time, I must admit, my life was definitely on the wane, at least professionally,” Stallone recalls. “I was emotionally, spiritually, physically, bankrupt. So I had $106, rent was $300, I had a pregnant wife—you might say I had quite a bit of fuel to sit down and write the script!” “But I felt that if I was going to go down, at least into professional obscurity, I wanted to at least have the opportunity to say to myself, well, you tried. You put your best foot forward and you didn’t make it. That’s what, I guess you might say, provoked me into writing the script, because it wasn’t something that I had given a lot of thought to. I just knew that I had to do it one time and it was innate.” “I wrote the script in about three and a half days and only because I could hear the wolf at the door, literally.” Most people would say it’s impossible to write an award-winning screenplay in three days. For Sylvester, the “wolf” broke his frame around how long things realistically take. But even though the wolf can be a powerful forcing function for speed, we can challenge our assumptions about how long things take without it. Opportunities to question our assumptions are present every day—all we need is the willingness to take up the challenge. via Action Digest

Ideas, Missives & Mentions

**Finally, here’s a set of ideas and worthwhile mentions (and stuff I want to keep out of web-scraper reach) intended for those I work with (free for founders in my portfolio, Adobe folks…ping me!) and a smaller group of subscribers. We’ll cover a few things that caught my eye and have stayed on my mind (including questions around AI Agents, a terrifying demo, as well as my latest areas of interest as an angel investor. Subscriptions go toward organizations I support including the Museum of Modern Art. Thanks again for following along, and to those who have reached out with ideas and feedback.