News
The Fed Just Ghosted the Market
The Fed just broke a 15-year tradition. And it tanked the markets.
What happened? On Wednesday, Kevin Warsh ran his first FOMC meeting as Fed Chair. This is the new Fed chair appointed by Donald Trump.
It was noteworthy for two reasons.
#1. Rates held at 3.50–3.75%
And it was a unanimous decision by all the members of the Federal Open Market Committee.
Most analysts were expecting it. But the "dot plots" were a surprise for many analysts.
The "dot plot" is the Federal Reserve's quarterly chart showing each FOMC member's individual forecast for where the federal funds interest rate should be over the next 3 years and into the "longer run".
And the dot plot turned hawkish: 9 of 18 members now pencil in at least one hike before year-end, 6 see two. There were 0 members favoring a rate hike in March. So, it's a big deal.
Further rate hike means less liquidity to pump risk-on assets like our crypto bags.
Kevin Warsh was seen as "Trump's guy" who would reduce the rates to pump the market. But the first meeting didn't go that way.
#2. Killed the Forward Guidance.
Forward guidance is a central bank tool that communicates its likely future course of monetary policy. It essentially tells markets where interest rates are expected to go.
For ~15 years, markets have been trained to trade the guidance, parse every word, and front-run the pivot.
Warsh just broke the script. The Fed is no longer promising anything. It means less predictability and a higher uncertainty baked into everything. Combine that with a Fed that's now signaling hikes (not cuts) into sticky 4% inflation, and the markets dropped. BTC fell ~2.2%, breaking $64,350 support toward $63K. ETH fell ~3.6%. Even gold dropped.
Why did Warsh kill "Forward Guidance"?
According to him, it is "not well suited for the current policy conjuncture". What does that mean?
1. It can become a fake promise the Fed can't keep.
With inflation still near 4%, the Fed genuinely doesn't know if it'll hike, hold, or cut. Promising a path it might have to break just burns credibility. Better to promise nothing than to promise wrong.
2. He wants markets to stop front-running the Fed.
When the Fed telegraphs moves, traders pile into that bet early, and that itself loosens or tightens conditions before the Fed acts. Warsh wants the Fed to react to data, not to a script it announced months ago.
3. It's a philosophy thing (the "regime change").
Warsh has long argued the Fed talks too much and micromanages markets. His view: the Fed should be a steady hand focused on price stability, not a play-by-play narrator. Stripping guidance forces everyone (including the Fed itself) to actually watch the incoming data.
The simple analogy
- Old Fed: a GPS constantly announcing "turn left in 2 miles." Traders memorize the route and crowd the exit early.
- Warsh's Fed: no announcements. It just drives, watching the road (the data) and reacting in real time. Less comfortable for passengers, but harder to mislead, and you can't blame the GPS when conditions change.
The tradeoff: less predictability and more volatility, because markets no longer get the cheat sheet.
Kevin Warsh also announced five task forces. They are: communications, balance sheet, data sources, productivity & jobs, and inflation frameworks. He said most should wrap up work by the end of 2026 and will "start with first principles, ask hard questions, examine current practice, consider alternatives, and propose next steps".
The productivity and jobs task force will specifically examine AI's economic impact. So theoretically, if the AI massively increases productivity, the Fed might run the economy hot. But it's just pure speculation for now.
Overall, there's a "regime change" in the Fed and how it runs the macro. If macro is interesting for you guys, we'll keep covering it.
Analysis
The Fable Ban: Does Crypto Actually Fix This?
Crypto x AI got a big narrative catalyst this week.
What happened? The U.S. government forced Anthropic to suspend all access to Fable 5 and Mythos 5 to any foreign national. It included every foreigner, whether inside or outside the United States.
So, just three days after their release, Anthropic shut down Fable 5 to everyone.
The US Commerce Department cited "national security threat". We don't know the exact "national security threat". But it seems like there was a way to "jailbreak" Fable and use it for harm, like hacking.
While it sounds bad for Anthropic, it was a big marketing boost for them. They can now claim to be the only AI model soooo powerful that US Govt banned it.
Once those jailbreak issues are patched, I expect Anthropic to open Fable again.
Many crypto people used this incident to promote the "crypto x AI" narrative. Many coins at the "AI x Crypto" intersection pumped.
- $TAO (Bittensor) rallied ~20% in the first 24 hours.
- $VVV (Venice) gained ~14% in the 24 hours after the news.
- $MOR (Morpheus) pumped as well. This was a sector-wide rally.
So was this a fundamental re-rating, or just a sentiment rally?
Imo it was purely sentiment. Nothing about these projects' actual capabilities changed in 12 hours. The story just got a perfect headline. Venice is the one name genuinely adjacent to the news, and it was already re-rated before the ban, printing a new ATH of $21.32 on June 3, ten days early.
When it's a narrative rally like this, the move tends to fade as fast as it spiked and behaves like short-lived momentum. A fundamental re-rating is what would justify actually holding. This was the former imo.
Could crypto have actually solved this? Could any crypto project have kept Fable 5 running for users after the ban?
No. Fable is closed source. Only Anthropic has the weights. Nobody else does, crypto included. And the crypto "inference" projects you'd expect to route around this, like Venice, ultimately lean on the same closed labs for models like Opus and Sonnet. You can't decentralize access to a model you were never given.
So step one isn't crypto at all. It's open weights. The good news is that there are real open models now: DeepSeek, Kimi, Qwen, GLM. The bad news, if you're a crypto maxi, is that crypto didn't build any of them. Those are open source models from non-crypto labs.
Here's where crypto actually earns its keep: the access layer. An open model only matters if you can run it, and most people can't spin up a 70B model on their own hardware. That's the real job, giving anyone permissionless, private access to open models without asking a centralized gatekeeper for permission.
Venice is the leading project here. Its whole pitch is uncensored, permissionless access to open models (it also proxies closed ones), exactly the use case the Fable ban put a spotlight on. Founder Erik Voorhees seized the moment, casting the shutdown as live proof of why Venice exists.
Could crypto go further and own the model training layer too? Bittensor's Subnet 3 (Templar) just completed Covenant-72B, the largest decentralized pre-training run on record: 72B parameters, trained by 70+ independent contributors on commodity GPUs over normal home internet, weights public under Apache license. It scored 67.1 on MMLU, edging past Meta's Llama 2 70B.
But let's be honest about where that sits. Covenant-72B is competitive with a 2023 model, not with the frontier. Decentralized training is closing the gap on yesterday's best, not today's.
So that's the real state of crypto x AI. We have genuinely good projects giving private, permissionless access to open models, and that's a real answer to the kill-switch problem the Fable ban exposed. What we don't have yet is competitiveness at the model layer itself. Crypto solves access and permissionlessness today. It doesn't solve the capability.
But for a lot of use cases, running a slightly weaker model that nobody can switch off is a trade worth making.
🚀 DeFi Catalysts
MegaETH distributed USDm rewards to the points program users. They also launched MOSS, their ecosystem wallet for builders.
Cap Money has completed the auction for the $CAP sale. It got 1002 individual bids at $106M in fully diluted value.
Zama, in Partnership with Morpho and Steakhouse Financial, is launching a Morpho vault for Confidential USDC (cUSDC).
Plasma has launched Plasma One, the all-in-one Neobank-like unified app, on iOS. The Android version will launch this month.
Ventuals, the Pre-IPO Market on Hyperliquid, announced its shutdown. They're joining another team building in the Hyperliquid ecosystem.
Humanity Protocol announced a full token migration and 1:1 airdrop of a new H token after the June 8 exploit.
📰 Industry News
Moody's Ratings has expanded its Token Integration Engine™ (TIE) to the Solana blockchain. Their credit ratings can be read by Solana apps.
Coinbase made a lot of announcements in the "System Update: Take Control" event. It ranges from Pre-IPO perps to Coinbase for Agents.
Kraken launches a CFTC-regulated perpetual futures product for US Traders. It has 16 contracts where you can trade 24/7.
Robinhood has launched Agentic Trading for all customers. All exchanges are moving in this direction.
Blockworks, one of the leading media platforms, has acquired Messari, which was valued at ~$300M in 2022. Reportedly, they only paid ~$10M for the company.
🐦⬛ X Hits
- New era of Fed policy.
- Introduction to PropAMMs.
- A framework for Crypto Neobanks.
- Data businesses are getting rewired in real-time.
- Rysk founder analyzes Vitalik's stablecoin proposal.
😂 Meme
Until next time,
Edgy
Today's email was written by Edgy and Yayya.
DISCLAIMER: I'm NOT a financial advisor. This content is for education and information purposes only. Crypto and DeFi are risky and speculative. Please do your research before investing.
Whenever you're ready, here's how we can help you:
- 🚀 The DeFi Edge Ventures - We identify, invest, and help amplify DeFi Protocols that positively impact the Crypto space.
You're receiving this email because you signed up for my newsletter. You can update your Preferences or Unsubscribe here.
600 1st Ave, Ste 330 PMB 92768, Seattle, WA 98104-2246
No comments:
Post a Comment