GM!
Today’s top news:
Crypto majors continue grind up, led by Bitcoin; BTC +2% at $76,480 Saylor buys $2.54B in BTC, 3rd biggest buy ever; 85% powered by STRC AAVE publishes full incident report; losses range $120M-$230M Curve founder pushes for shared security standards across DeFi Polymarket raising $400M at $15B valuation, behind Kalshi’s $22B 🟠 Saylor Just Made His Third-Largest Bitcoin Buy EverSaylor’s Strategy now holds 815,061 BTC acquired for $61.56 billion at an average cost of $75,527. With BTC around $76K, the entire treasury is back in the green.
STRC did the heavy lifting again this week, as $2.18 billion of the $2.54 billion came from STRC preferred stock sales. And notably, this is the week Saylor also proposed moving STRC dividends to semi-monthly.
The thesis: a more stable STRC price means the stock trades closer to par for longer, which means more issuance capacity, which means more Bitcoin. And with Saylor now even 15% of the way through his latest $50B ATM, expect the Bitcoin buying to continue. In size.
Key Details:
Strategy purchased 34,164 BTC for $2.54 billion last week - third-largest single-week purchase in company history; holdings now at 815,061 BTC at an average cost of $75,527; BTC yield 9.5% YTD The funding: $2.18 billion raised via STRC preferred stock sales; $366 million from MSTR common stock; 85.7% STRC-funded Where things stand: 815,061 BTC at ~$75K = roughly breakeven on $61.56 billion invested; Saylor’s stated goal is 1 million BTC by end of 2026; $1.62B in STRF capacity and $26.73B in MSTR ATM capacity still availableBitmine’s total holdings are now at 4.976 million ETH, representing 4.12% of total circulating supply, and is now 82% of the way to its self-declared “5% alchemy” target.
Tom Lee commented: “While many believe the crypto winter may last through the fall of 2026, our view remains that the crypto winter is much closer to ending.”
He cited ETH being up 41% from its early February lows as evidence, and called it “the best wartime store of value.” He also noted ETH benefits from two tailwinds simultaneously: institutional tokenization flows and demand from agentic AI systems that need a neutral public ledger.
We will see if that bull case indeed plays out for ETH the token, but at least Tom is doing his part to help.
The round could grow to $1 billion with additional strategic investors beyond NYSE parent ICE, which completed its $600M tranche last month bringing total ICE commitment to $1.6 billion.
Notably, Kalshi just raised $1 billion at $22 billion, nearly 50% higher, and leads on both volume ($13B vs Polymarket’s $10.57B) and revenue (estimated $1.5B annually). Historically, these 2 prediction market giants have conducted raises in line with each other. This would be the first time Kalshi seemingly has pulled ahead. Apparently, Kalshi being the US regulated body (vs Polymarket still not live in the US) is the differentiator.
Here’s how the exploit went down: the attacker forged a LayerZero packet (nonce 308) accepted by Kelp’s single-DVN bridge with no corresponding burn on Unichain releasing 116,500 rsETH. Then 89,567 of that ended up on Aave as collateral across 7 wallets, which borrowed 82,650 WETH and 821 wstETH against it. All 7 attacker positions currently sit at health factors between 1.01 and 1.03 and haven’t been liquidated yet.
The key problem is that the adapter that backs all remote-chain rsETH across every L2 currently holds only 40,373 rsETH, against total remote claims of 152,577. That’s a 112,204 rsETH gap.
How Kelp socializes the losses determines how bad this gets for Aave. If the haircut is spread across all rsETH holders globally, the token depegs roughly 15% and Aave absorbs around $124M in bad debt. If losses are isolated to the affected L2 chains instead, the impact concentrates on Arbitrum and Mantle and Aave's bad debt swells to roughly $230M.
Aave is running scenarios but can’t close the book until Kelp makes a public decision. Thus the holding pattern continues, and unfortunately, confidence in these protocols is seemingly dropping by the hour…



















