Planning: Working backwards
We all talk about working backwards – as leaders, good stewards of business, and high-class planners – but for the first time, the world of AI is talking about that with Q-Learning. This year, we will hear a whole lot more about machines learning to plan!
Deep learning + Planning
Deep learning will rapidly fuse with reinforcement learning this year, as researchers and developers alike move past the simple “next token” tinkering. AI models will need to work with strategy, goals, discounted rewards, and… “working backwards” planning. Otherwise we will continue to struggle with hallucinations, chain-of-thought prompting, and naysayers saying “I told you so.”
China plus one, OpenAI plus one planning
Of course every company worth their (manufacturing) salt is thinking about supply chains beyond China, the in-vogue “China plus one” strategy. Startups and software enterprises will be thinking hard about their own AI supply chains, and whether OpenAI alone will suffice. Because video and images will start to play a much bigger role in the enterprise in 2024, the question is whether open source diffusion models will be the way that the enterprise hosts its own models in their own environment. Text is probably cheap enough to call OpenAI for. What about video, image, and voice?
Global talent planning: All things distributed
2024 is a fork in the road for two kinds of companies – those that will continue to stay remote as they optimize EPS and bottom lines, and those that will bring people back into global clusters as a compromise between large office campuses and no office at all. Like most things in life, the pendulum does not rest other than in the middle! For those continuing to stay remote, college hiring and young talent development will make it extremely difficult to continue to not have offices (“watering holes”).2024: A Year of Rewards, Herds, and Valleys 2024: A Year of Rewards, Herds, and Valleys. Thundering herd euphoria: Is it “Game Stop” time yet?
While I don’t think we are back to stock market speculations driven by subreddits of 2022 anytime soon, there will be a euphoria like there was one in late 2009 to early 2010, when the Nasdaq Composite went up almost 40% in 8-10 months. I am speculating that the markets react like a thundering herd, even on the slightest 25 basis point interest rate reduction by the Feds.
VCs have lost a lot of money in the past 3 years of the {crypto | covid | ai} froth; so even while interest rate climate improves, VC as an asset class continues to be challenged by LPs expecting returns, which they can find in the imminent euphoria of the public stock markets.
Another bull that will continue to remind us of a thundering herd is India. It’s an incredible story. Reminds me of the $4 trillion China of 2006-07. If I were to make a wager, India does better than China of ‘07, because it can do it without the overspend on physical infrastructure – the hack that China used to pull forward growth – and the sobering slowness of the Global Financial Crisis of ‘07-‘09.
Yes, it’s a democracy. Yes it has the “federalist” slowness. Yes it has to deal with climate change and lack of energy security. But it’s all about the momentum. And by golly, what a sentiment of self-belief this country has at present – from enterprise spend to {zoom | upwork | deel}-powered knowledge worker momentum to government infrastructure and consumer spend, this force is unstoppable, at least for another 3 years.
The valley of despair: Digging out of it
VCs, startups, and the money scene
VCs are hurting. Consequently, startups continue to hobble along, many without a real runway that can see them past 2024. 2024 is the new 2003 (2011?) – investors struggling to raise new funds, and waiting for better times. Companies that we will see on the other side will have (a) conserved cash ruthlessly, (b) raised enough money and as a part of their compromise with their boards, retrenched and bought more time, or © raised money at discounts or in down rounds.
GPT wrappers and the trough of disillusionment
Many chatbot companies will disappear in 2024, as the market negotiates that it only has capacity for so many; most will try pivoting into Customer Support and Customer Experience, only to realize that it is a taller order than they had raised money for – building a system of record, content management, workflow engine, analytics database, copilots, search, integration (marketplace), user monitoring, post-sales customization, etc. require 3-5 years of platform development and another hundred million dollars.
Product led growth, but with a tweak
PLG slowly makes a comeback, although in a different form than the cheap money era of “growth at all costs” inside sales, 1-year deals (throw the hot potato to customer success), “who cares about churn” methods, and airport ads as if there-is-no-tomorrow drunken revelry. Instead, it is now about product design – needle-in-a-haystack free tenants who carefully swipe a credit card because they see value – and a grounded community driven by meetups, APIs, webhooks, and marketplace developers. Reality check in 2024.
China: Buying growth and now reflecting
And finally China. There is so much that needs to be parsed here, all the way from the about turn of their zero tolerance COVID lockdown to the real estate meltdown to bad loans to investor and consumer sentiment. China will need to dig itself out of it in 2024, but this time around, that recovery won’t come from building bullet trains to nowhere. I predict that 2024 continues to be a difficult year for the country.
The Black Swan
So what is going to be the hardest to predict, and could be the most consequential event of 2024? The elections in the United States, and its implications on the Ukraine-Russia war. Both of these are known unknowns, but they’re intricately coupled, and will have a massive impact on human lives, the European economy, Ukraine’s sovereignty, and… Chinese ambitions (in Taiwan).
2024, in many ways, will define this coming decade of humankind, given how much is at stake 🤞
