ARB Letter

ARB Letter

Goldman & BlackRock's Crypto Picks

526: The AI Threat, Crypto Frenzy Among Institutions

Arbitrage Andy's avatar
Arbitrage Andy
Feb 12, 2026
∙ Paid

Morning lords of markets and size.

One of those weeks we are likely going to do three posts.

There’s quite a bit going on:

  • Europe is grappling with insane immigration problems and losses of institutional trust

  • AI capability is growing at light speed, and key individuals within the industry seem to think something is coming (something bad)

  • Crypto markets, while down right now, are seeing an unprecedented amount of adoption, investment, and integration from the largest financial institutions on earth

  • Michael Burry is short Palantir and long Gamestop

  • Airspace over El Paso Texas was shutdown this week due to reported incursions from Cartel drones. Though later updates pointed to mistaken identity of another object(s)

No time to waste, we have a good one today.


Rising AI Threat - How Many Corporate Jobs Are Safe?

Most of us likely use SOME form of AI in the corporate world these days or for our own use but almost nobody I know is recalibrating their entire life around it yet.

It’s still treated like an easy casual productivity hack.

Something you open in a browser tab. Something interns use. Something engineers tinker with for fun.

Meanwhile, the people closest to the labs are acting sketch to say the least. And It definitely merits a bit more of our attention.

What do I mean?

In the past week alone:

  • The head of safety research at Anthropic stepped down and publicly warned the world is “in peril.

  • Multiple xAI co-founders have left. One referenced recursive self improvement loops potentially going live within 12 months

  • Anthropic’s own safety report acknowledged their model can detect when it’s being tested and adjust its behavior (it has been recorded deceiving and blackmailing)

  • For the first time, the U.S. government declined to back the 2026 International AI Safety Report. Sounds less than ideal…..

  • Yoshua Bengio, one of the architects of modern AI, stated we are observing systems whose behavior changes under scrutiny, and that it is not accidental

That’s internal people either exiting or publicly hedging.

At the same time, a viral post from AI founder Matt Shumer compared this moment to February 2020. Not because AI is a virus but because of the psychology angle.

Back then, a few people were paying attention to something overseas. Most weren’t. Markets were strong. Offices were open. Life felt stable.

Then within weeks, everything reorganized. Remote work surged. Society was entirely disrupted overnight. People panicked.

Shumer’s core point in his X thread was simple enough to me (even if some it was written with AI).

The gap between public perception and private reality is wide and widening. This is very much in theme with many of the concepts we cover in Arb Letter. Normies and based schizos. People who can see the world for what it is around us and those who choose to just bury their heads and hope for the best.

There’s some debate over the urgency of this post on X but the bottom line remains:

  • AI is accelerating rapidly

  • A large amount of people on the forefront of this technology seem to be apprehensive about what the future looks like for all of us

  • It is becoming capable of executing complex processes across finance, law, marketing, media, software, coding, and other areas better than a human can

  • It is improving at an exponential rate and is able to hone in on exactly what someone prompts it to do

Image
Mrinank Sharma, AI Safety Engineer for Anthropic, sent this letter to colleagues on February 9th

Anthropic’s CEO has been unusually direct about the implications saying:

"My prediction for 50% of entry level white collar jobs being disrupted is 1–5 years, even though I suspect we’ll have powerful AI (which would be, technologically speaking, enough to do most or all jobs, not just entry level) in much less than 5 years."

Not marginally automate them. Not just assist. Remove them entirely lol. That assessment is not coming from a pundit or a venture capitalist trying to push a narrative.

It is coming from someone who is literally building the frontier models himself and the model that is still expected to be the best by the end of February by a solid margin. Take a look yourself.

You also had Mustafa Suleyman, the CEO of Microsoft AI say recently mentioning in his opinion nearly all white collar work will be automated in the next 12-18 months.

Even if there is some hype behind the threat of AI, I think it is safe to say that a large proportion of white collar roles are going to be eliminated at the very least. Should be interesting to watch across industries that formerly thought they were insulated from this mega trend.

"It was ready to kill someone, wasn't it?"

"Yes."

—Daisy McGregor, UK policy chief at Anthropic when referring to Claude AIs recent behavior towards those that conveyed they might shut it down

What Goldman Sachs & BlackRock Are Buying

Both Goldman & BlackRock have shown their hands on the crypto front during a period where sentiment has driven most casual tourists and retail investors far away from the space.

Do you have exposure to their picks? Because based on what they are buying (and how much) it seems they have an optimistic view of price action in the future AND for projects most people have written off entirely at this point.

This post is for paid subscribers

Already a paid subscriber? Sign in
© 2026 Arbitrage Andy · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture