Hey people! Welcome back to Forests Over Trees, your weekly tech strategy newsletter. It’s time to zoom-out, connect dots, and (try to) predict the future. Let’s go!
Did anyone else learn to square-dance in highschool?
In 9th grade in Atlanta, it took up multiple weeks of gym class. Weeks! That’s a lot of time. Time that could have been spent playing dodgeball, over-using Axe body spray, etc. You know… important 9th grade stuff.
But rather than a dodgeball whizzing towards your face, the panic-inducing part of square-dancing was the partnering-up. You never knew who you’d have to dance with.
Maybe your enemy. Maybe your crush. Maybe your enemy’s crush.
And with square-dancing, everything constantly changes. You’ll partner up, dance for 30 seconds, and then boom – on to the next dance partner. It’s a cycle of panic, partnering, relief, and then panic again.
This week in tech, I felt like I was back in 9th grade.
AI companies are chaotically, frantically partnering up. So today, together, we will try to tame the chaos and make sense of it.
But first, a quick reminder about the power dynamics going on here.
Sources of (AI) Power
Data & media - they have the raw materials used to train AI models (ex. social media companies, news sites, etc.)
AI platforms - they train the AI models, turning raw materials into complex formulas of variables and weights.
Cloud providers - they rent-out compute clusters needed to train and run models.
Chip-makers - they produce the hardware that goes in the compute clusters.
I’m intentionally using bullets here instead of a numbered list.
Why?
Because there’s no single “most important” piece. You need all the pieces.
However, you don’t necessarily need to own all the pieces. You can partner. Hence the chaotic partnering up…
So Who’s Dancing?
Great question. Complicated answer. Some folks are only dancing with a single partner, some folks are partnering with themselves (going vertical), and some are contracting with multiple partners.
Here’s an example of each:
Make sense so far? Ok good.
Now, one element of the chaos here is that almost everybody is moving from standard or vertical strategies into the multi-partner approach. For example, rather than just Tumblr in the Standard/Data cell, it’s actually Tumblr + Shutterstock +…
The benefit of the multi-partner approach is that you can move faster – joining forces with specialists in areas where you aren’t the best. You sacrifice control and cost in exchange for speed.
The shift in strategy means a lot more partnerships are getting announced, which should not be surprising! Let’s talk through a few recent ones.
The Partnership News
Mistral <> Microsoft – Mistral is a Paris-based AI company, and people were SHOCKED to hear Microsoft signed a $2.1B deal with this new (non OpenAI) partner. But it makes sense if you think about it. When all the Sam Altman drama happened, Microsoft realized they had a single point of failure with their AI play. You can’t leave that type of risk un-mitigated, even if: a/ Sam could have followed through and re-created OpenAI within Microsoft or b/ Microsoft could have recreated the models on their own. There are too many if’s there.
Tumblr <> OpenAI – Leaks about this started earlier in the week, so we don’t have all the details yet, but Tumblr will sell data to OpenAI for training. And the news headlines I saw were framing this as a bad thing. Let’s be honest, people, anyone who’s able to turn shit into salad and salvage value from Tumblr should be applauded, not criticized. Plus, Matt Mullenweg, the founder of the company that now owns Tumblr, is one of the good guys in tech (as I’ve written about before).
Reddit <> Google – People were poo-pooing this one because Reddit will “only” get $60M per year. But Reddit probably wanted to announce some good news before their March IPO to get the good vibes going, and they’ll get some help with a more powerful search tool from Google. Plus, Reddit was handing away free data to OpenAI for training a few years ago (see more here)… some money is better than none.
Wrapping-up
“So what?” I hear you asking.
Well, the larger point is that every name I’ve already mentioned, and any of the other new ones that might pop up in the next few years… they are jostling for position.
AI is a chance to re-invent your company. We are in the chaotic, “anything could happen” stage of the game right now.
French startups you’ve never heard of are getting billions from Microsoft, who used to be the boring “I’m a PC” of tech. Nvidia’s stock is dizzyingly high. People are treating Google (GOOGLE!) like a has-been. These are crazy times.
So while that chaos unfolds, expect more wild partnerships. Expect to be reminded that a company exists when you hear that they’re selling your data. Expect weird strategies.
The tech square-dance is in full swing.
Bonus Bullets
Quote of the Week
“I’m skeptical of maintaining the status quo. Technology, on average, makes things a lot better and leads to more wealth across the board. It has to be managed appropriately, but I don’t think it should be stopped.”
– Vlad Tenev, Robinhood CEO (from an awesome Semafor interview)
Quick News Reactions
Block stock – it jumped ~23% in the past week. In December, I wrote about how Jack Dorsey is doing the equivalent of 100 pushups a day (to convince us Block is getting fit and is now in profit-seeking mode). But really they’re just doing the same thing that other tech companies are doing (grinding earnings up, laying-off people, promising more cost-cutting, etc.). Still though, not a bad start to their 2024, financially-speaking.
Japan welcomes TSMC – The chip-maker will build plants ($20B+ planned investment) and get subsidies. Just like with the TSMC-US deals, TSMC won’t build their most advanced versions there, but it’s a start (and a good move) for Japan. They’re a world-leader in materials for chips, so it builds on a strength and rides the wave…
Klarna drama cont’d – Last week, I mentioned the VC tumult around Klarna.. this week, we found out a likely reason why: a planned Q3 2024 IPO. The VC’s are jostling for position ahead of the potential payday.
Overall Economy
This is the Weekly Economic Index published by the Dallas Fed. It’s made up of 10 different data sources from consumer to labor to production, and it’s designed to closely track US GDP.
Tech Equities & Bitcoin
The Nasdaq (blue) closely tracks tech equities, and I added the S&P 500 (green) and Bitcoin (orange) for comparison. Note: this is not investment advice, but it is interesting.
Tech Jobs Update
Here are a few things I’m paying attention to this week.
Big Tech Job Posts: LinkedIn has 11,298 (+71.5% WoW) US-based jobs for a group of 20 large firms (the ones I typically write about — Google, Apple, Netflix, etc.).
Layoffs from 2022-2024: (Source: Layoffs.FYI). Note that this is showing in-progress numbers for the current month.
