Thoughts about the markets, automated trading algorithms, artificial intelligence, and lots of other stuff

  • The Jobs Most Impacted By AI

    As we talk about the proliferation of AI, it's probably helpful to see where it's predicted to have the most impact. 

    Job_Departments_Impact_by_AI_sitevia visualcapitalist

    These results come from a World Economic Forum report

    In context, large impact refers to full automation or significant alteration. Small impact refers to less disruptive changes. 

    IT and finance have the highest share of tasks expected to be "largely" impacted by AI … which is unsurprising. 

    We've also already seen the impact of LLM and generative AI on customer service and customer care. As these tools improve, more cases will be able to be fully handled by AI. 

    This chart isn't meant to make you feel afraid that your industry will be automated—it's meant to help you understand what tasks you should consider automating. 

  • Applications Of Data Analytics & AI For Your Business

    It's a common theme in entrepreneurial discussion these days … AI is coming for your jobs. 

    The more nuanced statement is that AI isn't going to take your job – but someone using AI better might. 

    Recently, Andrej Karpathy, ex-director of AI at Tesla and founding member of Open AI, posted a great tweet about how software engineering will be automated.  He compared it to automated driving. 

    With automated driving:

    1. first, the human performs all driving actions manually
    2. then, the AI helps keep the lane
    3. then, it slows for the car ahead
    4. then, it also does lane changes and takes forks
    5. then, it also stops at signs/lights and takes turns
    6. eventually, you take a feature-complete solution and grind on the quality until you achieve full self-driving.

    The progression is similar for software engineering (and, you guessed it, your business as well)

    1. first, the human writes the code manually
    2. then, GitHub Copilot autocompletes a few lines
    3. then, ChatGPT writes chunks of code
    4. then, you move to larger and larger code diffs
    5. then, a tool starts coordinating other tools (a terminal, browser, code editor, etc.)

    You get the point.  Human oversight begins to move towards increasingly higher levels of abstraction and management. 

    If you think about it, this parallels a pretty generic path that a typical employee might take in your business.  A junior employee can't handle any ambiguity.  As they move up, a mid-level employee can probably handle some mild ambiguity … they need to know where they're headed, but they don't need hand-holding on how to implement it.  A senior employee needs to know what problems they need to tackle, and then you get to entrepreneurs, and they don't even need to know what problems to tackle … they'll find some. 

    Evolution

    This suggests a pretty solid modus operandi for the coming years.  If you're worried about being replaceable, focus on higher-level behaviors

    AI empowers businesses to do more with less.  Early adopters of AI will gain a significant competitive advantage by automating tasks, enhancing customer experiences with personalized recommendations, and making data-driven decisions that lead to cost savings and increased revenue.  Integrating AI into your business will propel your organization forward by unlocking new levels of efficiency, effectiveness, and certainty.  If you're steering the ship, you don't need to be as afraid of the waves. 

    Here is a framework I created to identify the path to some not-so-easy wins that lead to sustainable business growth and progress: 

    • Create Process Playbooks that leverage automation and AI to help businesses exceed standards both front-stage and backstage.  This class of solutions improves practical and business outcomes and helps avoid errors, omissions, and discretionary mistakes.
    • Use Outcome Integrity Trackers to log decisions, actions, and results, hopefully improving and standardizing processes and outcomes.  This capability will evolve into the ability to measure the difference between skill and luck reliably and to the creation of accurate recommendation engines with real-time expectancy scoring.
    • Capture, Calculate, and Curate Custom Metrics.  Much of what happens each day is lost.  Finding a way to save this data creates, expands, and augments a valuable new asset that is valuable itself, helps solve complex problems, and leads to new products, services, and solutions.
    • Curate a Single Integrated Source of Trusted Data that is accurate, complete, and up-to-date.  Together, that data becomes the foundation for building new models, metrics, validations, certification, and compliance solutions.

    Developing a Comprehensive AI Strategy is Crucial for Business Success

    Businesses that don't adapt to changing landscapes fail. Having a roadmap, centered on what doesn't change is a reliable life support. Change doesn't have to be dramatic to be valuable. Just by taking these little steps and asking the right questions, you can make a big impact. I hope you're finding way to reap the rewards of these transformations, not just surviving them. 

    Message me if you want to talk more about this.

  • Overhyped Technologies (Or Not)

    Just because something is overhyped doesn’t mean it’s bad.

    Gartner’s Hype Cycle is a great example of this concept.  It highlights the likely cycle of inflated expectations, disillusionment, and, ultimately, utility.

    The key takeaway from the Hype Cycle model is that much of what happens is predictable … and that a significant portion of the extreme swings are based on human nature rather than technical merit.

    Haters are going to hate, and sometimes a fad is more than a fad.  For example, here is a front-page article from the New York Times in 1879.  It questions the utility of electric lights as a replacement for gas-powered lighting.  In case you were wondering, that one might have been a bright idea.

     

    Screen Shot 2022-05-15 at 8.45.33 PM

     

    The point is that humans have proven themselves to be pretty bad at exponential thinking.  We’re not bad at recognizing periods of inflection, but we often have trouble recognizing the consequences of the change (and the consequences of those consequences) and predicting who the winners and losers will be as a result of those regime changes. 

     

    Screen Shot 2022-05-15 at 2.26.23 PM

     

    There are countless examples.  Here’s a funny one from Maximum PC Magazine in 2008.  It shows that hype isn’t always a sign of mistaken excess.  This list purported to show things that were getting too much attention in 2008.  Instead of being a list of has-beens and failures, many of these things rightfully deserved the attention and hype they were getting.

     

    Maximumpc

     

    It’s been over 15 years since this came out.  How did the predictions hold up?

    Apple has become one of the world’s biggest and most successful companies (with a market cap approaching 3 Trillion dollars).  The iPhone has sold over 2.2 billion phones and accounts for over half of Apple’s total revenue.  Meanwhile, Facebook has become Meta and is also one of the biggest and most successful companies in the world (with a market cap of well over a Trillion dollars).  And the list keeps going: HD video, 64-bit computing, downloading movies from the internet, and multiple GPU video cards. 

    Take just that last one. Nvidia has been the primary beneficiary of GPU growth, and it is one of the highest-performing stocks of the past few decades (with a market cap of well over 2 trillion dollars). 

    It’s hard to believe how poorly this image aged. 

    Remember that the trend is your friend while it continues.

    Just because something is overhyped – doesn’t mean you shouldn’t be excited about it

    The key is to stop thinking about the thing that’s being hyped and, instead, to start thinking about how to use things like that to create what you really want.

    Onwards!

  • Velocity Versus Speed

    Recently, I've been thinking a lot about how businesses scale and technology adoption accelerates.

    For example, consider how fast AI is improving and transforming business.

    Last year, I shared a short video called Speed Matters. It includes some thought-provoking ideas. You can view it by clicking here.

     

     

    While speed matters, faster is not always better.

    As you focus on doing more things faster, it becomes more essential to give people room to do important things slower.

    You've almost certainly heard the phrase, "It's better to measure twice and cut once."  It's much easier to do something the right way from the beginning rather than trying to fix it after you mess it up. 

    Activity does not create progress if it doesn't move you in the right direction.

    This reminds me of a distinction my friend Nic Peterson makes. What you want in your business is velocity rather than speed. Velocity implies a vector (and preferably only one vector). You want to move fast in the desired direction, not fast towards distractions, mistakes, or money down the drain. 

    To add one more layer to this, there's an "almost true" axiom in technology: you can only have two out of these three things: a project done fast, done right, and done cheaply.

    But it's only almost true for a big reason. If you've already built the team and put in the work, replicating that work for new projects can be done fast, right, and comparatively cheaper. 

    Are you just moving fast, or do you have velocity in your business?

    Something to think about … 

  • House Prices Versus Income in America

    As we discuss the economy, I also think about my youngest son, who is looking at houses right now. 

    In countless ways, today's youth have it easier than we did. Access to opportunities, the internet, capital sources, etc., has gotten more accessible, yet there are a few things that have gone the other way, such as buying a house. 

    OC-U.S.-Income-Housing-Gap_Feb14

    via visualcapitalist

    The chart above does not show interest rates or inflation. For example, in 1984, the 30-year fixed rate was close to 14%, over double what it is now. 

    But, to put things in perspective … I moved to Texas in 1986. Part of my rationale was that I could buy a "nice" home for a little less than my initial starting salary as a lawyer.

    Recently, policy decisions have vastly increased house prices. How much? Median house prices are nearly 6x the median household income in America. Meanwhile, the economics of renting are significantly better than buying. According to the WSJ, it's 52% more expensive to buy than rent due to mortgage prices

    When housing costs are this high, consumer spending and mobility are reduced, making individuals less likely to relocate for job opportunities. 

    We live in interesting times. Sometimes, I miss the good old days.

  • Checking Back In with Dr. Doom

    In 2018, while in New York for work, I was invited to a party that ended up being a pretty unique experience. 

    The rules were that for the first hour it would be first names only, no discussion of what you do, etc. Part of the fun was figuring out who was there and why they were special. And, there were a lot of pretty impressive people in the room. As I was wondering who was able to bring all these experts and thought leaders into one home for a house party, I found it was Nouriel Roubini – the infamous Harvard economist known as Dr. Doom. 

    Nouriel Roubini's predictions have earned him the nicknames "DrDoom" and "PermaBear" in the media. He predicted the housing bubble crash in 2007-2008, and has extensively studied the collapse of emerging economies. 

    So, after a tumultuous few years for the global economy, I thought I'd check back in and see what he was saying. 

    It turns out he's less pessimistic than you would guess. He's pretty optimistic about 2024 growth and not particularly worried about a recession—though he is expecting a downturn. 

    He also thinks there's a possibility that growth remains above potential, and inflation remains sticky. That would be good news for the economy, but bad news for markets – as the Fed likely wouldn't cut as much or as soon as people are hoping for. 

    Now, Nouriel has been wrong before, and I don't trust any singular pundit. My mindset is to listen to voices that don't already believe what I do. I tend to be optimistic as a rule, and I've been optimistic on things like blockchain, whereas Nouriel has been staunchly negative.

    But, he's a smart and educated voice who can justify his opinions. And I end up more educated – and often modifying my stance a little bit – based on the context he's able to give. 

    By the way, as I was editing this post, I saw that Chase CEO Jamie Dimon and billionaire hedge fund founder Ray Dalio admit they got warnings on the US economy wrong — for now.

    Are you listening to voices outside of your preferred channels?

  • The State Of Pop Culture As An Allegory For All Culture

    There are talks of a Gladiator 2, of remaking American Psycho, of another John Wick or Fast & Furious

    It feels like art has begun to repeat itself more than before. That's not to say we've never had copycats or derivative works … but that as media has become more ubiquitous, its influence has become more heavy-handed. 

    It feels like there isn't a place in the current market for mid-budget films targeting a niche market and a release that makes a bit of money… Instead, it's all blockbusters or straight-to-DVD (now straight to streaming.) 

    While it "feels" that way, the data agrees with that belief. 

    2ecc7ddd-23d6-410e-909e-32cec48a23c7_600x380

    via Adam Mastroianni

    I'm using movies as the benchmark, but if you go to Adam's article, You can see it's true for authors, musicians, the silver screen, and more. 

    Media, like the news, and capitalism, has arguably become an oligopoly. 

    It's interesting to see because, in all three scenarios, the internet has led to the rise of indie initiatives. There are more entrepreneurs than ever. There are more independent news sources than ever, and by a large margin, more "art" is being created than ever before. 

    Access to ears, creation, and opportunity has never been more widespread.

    And yet, the vast majority of dollars and eyes have congregated on a select few. Like the dwindling middle class, art seems to be moving away from the mid-market. 

    Mastroianni offers various theories, but I'd love to hear yours … 

    What do you think is causing this shift, and what do you think it would take to create a more equal distribution?

    Is it even possible? Does human nature prevent it?