Matt @ Meritum retweeted
Fantastic 5-part series on insurance brokers from @MeritumInvest. Well worth the read.
Curious if others agree with his AI vs. cycle risk exposure framework (attached).
And here's where Forward P/Es stand today:
$BRO at a discount (~15.5x)
$AJG / $RYAN at a premium (~19.6x and ~19.9x)
Which broker would you pick?
$MRSH $BRO $AJG $RYAN $AON
Wonder how Anthropic IPO bankers are feeling with..
1) terrible earnings reactions in AI,
2) anti-AI democrats set to sweep,
3) OpenAI terrible raise,
4) Hyperscalers soaking up all capital in market with strategically placed raises.
Yet nobody doubts IPO gets done?
1
47
We have released a short and digestable summary of our long thesis on $RELX - our largest position for a few months now.
meritum.substack.com/p/relx-…
123
Anthropic indicated "ARR" of 65bn at end of Q2. By my math that comes out to june being a 5.4bn month.
If 0% growth, anthropic should then have reported 16.3bn rev in Q3 (5.4 x 3).
Instead it was 16.8bn? So 3pc growth from end of june starting ARR implied no growth?
Am i missing something, or is this terrible?
Today we cover Nordrest ($nrest.sto / $nrest).
Good company, but likely overextended on margins and potentially a levered bet on European MRE stockpiling, which may end sooner than Ukraine does.
meritum.substack.com/p/all-s…
Also important: What do we think the depreciable life of a model is? How much cash flow are models from 2024 generating today? I think model lifetimes are 1-2 years, almost expensable.
So if you are forced to reinvest in R&D every year, it should just be treated as an expense.
The accounting trick is simple;
1. you buy GPUs (recorded as asset)
2. you train model (recorded as R&D)
3. you deploy model (revenue)
4. you pay for power & cooling (cost of revenue)
The S1 only shows you 3 & 4
(80% margin omg!)
How do you get 3 without 1 & 2?
1
81
Matt @ Meritum retweeted
The accounting trick is simple;
1. you buy GPUs (recorded as asset)
2. you train model (recorded as R&D)
3. you deploy model (revenue)
4. you pay for power & cooling (cost of revenue)
The S1 only shows you 3 & 4
(80% margin omg!)
How do you get 3 without 1 & 2?
Matt @ Meritum retweeted
this is Dario and sama's worst nightmare -
A law firm buying Nvidia servers.
Latham & Watkins is building an in-house AI stack..
this is US’s second-largest law firm with $8.3 Billion in revenue last year
And now it has -
- Nvidia hardware it controls
- open-weight models it can fine tune
- proprietary legal data it is trusted to protect
- infrastructure only Latham employees can access
A law firm has decades of contracts, negotiations, client context, legal reasoning, and institutional knowledge.
They dont want to give all of that away to OpenAI or Anthropic in exchange for expensive tokens..
And on top of that - risk their data being used to train frontier models..
This will happen more and more now..
The big AI labs have no moat.. nothing protecting their largest customers from moving on..
The biggest companies in the world will -
- own the compute
- own the data
- own the workflow
- fine tune the model around their business
- switch providers when the pricing or quality changes
Dario and sama want to make this illegal by bringing in regulation.. and become the AI overlords..
Matt @ Meritum retweeted
Replying to @SqueezeMetrics
"Look at him, framing questions about depreciation, about how fast these computers are written off. I know what he's going to ask. He's going to ask what makes a finance company worth fifty times earnings. Right?"
Matt @ Meritum retweeted
"He actually did see the things about to happen"
"We were on the verge of [...] technical breakthroughs that would change the ways of living"
"Engineers would come to Harry's apartment and tell him what was going on"
"You have to, or you end up [in the permanent underclass"
$AON acquires USI for 17bn - "tenth largest U.S. insurance broker with approximately $3 billion in annual revenue"
So around 5.7x sales, and approximately is only used when its just below... probably +20x earnings.
Looks expensive to me... pretty much above the trading range of every broker over the last 5 years. Why not just do a big SBB?
1
2
751
Name one single company where this has occured at +1bn scale? I.e. somebody with a major lead gets disrupted by a tokenmaxxxoor
Ben Horowitz on the oldest law in startups, and why the industry no longer has to obey it:
"The one thing we all knew in startup world is that if I have a two-year lead on you and you try to catch me by hiring 1,000 engineers, you're gonna wreck your company."
"That never works. It's a Mythical Man-Month. Nine women can't have a baby in a month. That's it. That never works."
"Okay, now that works. But it's not hiring 1,000 engineers. It's taking $3 billion and lighting up a magnificent cluster, and then all of a sudden, whatever, Grok can come out of nowhere and, oh, all of a sudden it's real."
"You can throw money at the problem, and you can throw money at almost any problem, and that works. That is just completely different than anything we've ever lived through. By the way, we're all psychologically adjusting to this."
@bhorowitz @eriktorenberg
138
+330% later... $AZO
Just a learning for myself & others... keep the long-term fundamentals top-of-mind, not peer multiples and quarterly SSS... In the long run EPS growth will take of you (AZO trades at same multiple today)
Apologies for the hindsight capital tweet, but just re-read an $AZO short thesis from last year. This guy nailed it.
1
226
Matt @ Meritum retweeted
Congratulations. If you own a hyperscaler, internet platform, internet marketplace, software, fintech or social media company, an asset-light financial services firm, a medical device provider, a consumer electronics business, you are again now the proud owner of an AI LOSER!
We are long $MBGL --- 13x normalized earnings for a strong duopolistic business with a decades track record of double-digit organic growth.
Strong consumer branding, contributory data models and used-car tailwinds provide for a stable, acyclical growth company at a bargain price.
2
4
643






















