A few things worth knowing today. MindRank just dosed the first patient in a Phase III trial for an entirely AI-designed obesity drug, and the timeline from zero to here is kind of remarkable. Northwell Health published results showing their AI tool cuts pancreatic cancer diagnosis time nearly in half. On the Bitcoin side, silent payments are getting refinements in the latest Optech newsletters, and a new protocol called BLISK is introducing boolean logic into spending conditions. Strategy bumped the yield on its STRC preferred stock again, now at eleven and a half percent, as Bitcoin pulls back a bit. And on the regulatory front, the digital euro is stalling in the European Parliament, U.S. stablecoin rules are tightening fast, and banks are actively blocking the CLARITY Act. Let's get into it.
Let's start with what might be one of the most concrete milestones yet for AI in drug discovery. MindRank, a biotech company, has dosed the first patient in a Phase III clinical trial for a drug called MDR-001. This is an oral GLP-1 receptor agonist targeting obesity and type 2 diabetes. If you've been following the GLP-1 space at all, you know this is the same class as Ozempic and Wegovy, except MDR-001 is a pill, not an injection, and it was designed almost entirely by AI.
The trial is called MOBILE. It will enroll about 750 patients across 50 clinical centers in China. It's led by Professor Linong Ji from Peking University People's Hospital. The earlier Phase IIb results were described as showing meaningful efficacy and favorable tolerability, though detailed numbers haven't been fully disclosed yet.
Here's what makes this genuinely noteworthy. MindRank used its proprietary platform called Molecule Pro to design this drug, and they went from project initiation to U.S. IND clearance in just 19 months. From start to Phase III in about four and a half years. For context, traditional drug development timelines from discovery to Phase III are typically ten to fifteen years. They've compressed that dramatically.
Now, the competitive landscape is brutal. Injectable GLP-1s have set extremely high bars for weight loss and blood sugar control. An oral formulation has obvious advantages for patient adherence and access, you don't need cold chain manufacturing and distribution, people are more likely to take a pill than inject themselves daily. But it also has to prove it can match or approach the efficacy of injectables, and gastrointestinal side effects are a known concern with this drug class.
The trial design is placebo-controlled, which will give clean efficacy data, but there's no active comparator arm. That means even if MDR-001 succeeds, its market positioning will depend on how the numbers stack up against existing treatments indirectly.
Meanwhile, also in healthcare AI, Northwell Health published a study in The Oncologist showing their AI tool called iNav reduced pancreatic cancer diagnosis time from 56 days to 35 days. It uses machine learning and natural language processing to scan over 10,000 MRI and CT reports per week, flagging high-risk cases for faster follow-up. Time to oncologist appointment dropped from 27 days to 17. Time from imaging to biopsy was cut in half, from 12 days to 6. For a cancer where early detection is everything, those weeks matter enormously. Both of these stories point to something real happening. AI in healthcare is moving past the proof-of-concept phase into actual clinical deployment where patient outcomes are being measured.
Let's talk about Bitcoin privacy, because a lot is happening at the protocol level that doesn't get enough attention.
The latest Bitcoin Optech newsletters, issues 392 and 394, highlight several developments. First, there's a proposal from Sebastian Falbesoner to limit the number of recipients in silent payment transactions. Silent payments, if you're not familiar, are a way to receive Bitcoin without reusing addresses and without the sender needing to interact with you beforehand. They're a significant privacy upgrade. But there's a theoretical attack vector where someone could create a transaction with a massive number of outputs targeting one recipient, which would bog down the scanning process. The proposed fix is a cap of 1,000 recipients per transaction group. It's a practical, sensible constraint that preserves the privacy benefits while closing an abuse vector.
Then there's a new draft BIP from Craig Raw that adds supplemental metadata to output script descriptors. This sounds very technical, but the practical impact is meaningful. It would let wallets annotate descriptor strings with things like wallet birthday dates and gap limits, which makes silent payment scanning more efficient and improves fund recovery. These are the kinds of quality-of-life improvements that make privacy features actually usable for normal people.
Also introduced recently is BLISK, a protocol by Oleksandr Kurbatov that enables complex spending conditions using boolean AND and OR logic. Current multisig setups with MuSig2 can only express basic conditions. BLISK allows much more sophisticated authorization policies while still resolving to a single signature verification key. This matters for both privacy and for building more flexible custody arrangements.
Stepping back, ForkLog published a useful overview of where Bitcoin privacy tools stand. CoinJoin remains the most established approach, combining multiple users' transactions to obscure the trail. PayJoin adds a twist by having both sender and recipient contribute inputs, which breaks the common heuristic that all inputs belong to one party. And silent payments represent the next generation of receiving privacy.
The broader picture here is that Bitcoin's privacy toolkit is maturing on multiple fronts simultaneously. It's not one silver bullet. It's a layered approach where different tools address different parts of the transaction lifecycle. The protocol-level work happening now, the BIPs, the proposals, the refinements, this is what makes these features robust and eventually ready for mainstream wallet adoption. It's slow, deliberate work, and it's exactly what Bitcoin needs.
Strategy, the company formerly known as MicroStrategy, just raised the dividend yield on its STRC preferred stock by 25 basis points, bringing it to 11.50 percent for March 2026. This is a perpetual preferred share with a variable yield that adjusts monthly, and it's designed to trade around its 100-dollar par value.
The timing is interesting. Bitcoin has been pulling back, and broader markets are jittery with macroeconomic uncertainty. So Saylor and CEO Phong Le are essentially sweetening the pot to keep STRC attractive to income-seeking investors during a soft patch.
This fits into a larger strategic shift we've seen from Strategy over the past year. They've been pivoting toward preferred shares as a primary funding mechanism for Bitcoin purchases, complementing their earlier approach of convertible notes and at-the-market equity offerings. If you look at where they were a year ago, they'd already raised over 18 billion dollars through stock sales and convertible notes, and they'd purchased roughly 195,000 Bitcoin. The shareholder meeting in early 2025 approved a massive expansion of authorized shares, both common and preferred, specifically to support their 42-billion-dollar capital plan.
The STRC instrument is clever in design. A double-digit yield on a perpetual preferred is genuinely attractive to a certain class of investor, particularly those who want Bitcoin exposure through a regulated equity instrument with income attached. And because the yield is variable and adjusts monthly, Strategy can calibrate it in real time to market conditions.
But there's a tension here. Every basis point of yield they pay is a cost against their Bitcoin treasury. They're essentially borrowing at 11.5 percent to buy an asset that might appreciate significantly but could also decline further in the short term. If you're a Bitcoin maximalist, you look at this and say the long-term math works because Bitcoin's trajectory over multi-year periods has consistently rewarded holders. If you're more skeptical, you see a company paying double-digit yields during a drawdown and wonder about the sustainability of that model.
What's undeniable is that Strategy has created an entirely new category of corporate finance. They are the template for what people now call digital asset treasury companies. Whether others follow their path depends a lot on how this next cycle plays out.
There's a lot happening on the regulatory front around digital currencies, and the themes are pulling in different directions simultaneously.
Start with the U.S. The CLARITY Act, which was supposed to bring regulatory clarity to stablecoins, has stalled again. The reason? Banks. Major financial institutions are actively blocking progress because they see stablecoins as an existential competitive threat. Stablecoins can do a lot of what banks do, hold deposits, facilitate payments, move money across borders, but without the overhead and legacy infrastructure. Banks are fighting to either kill the legislation or shape it so heavily that it neutralizes stablecoins as competitors.
At the same time, the Office of the Comptroller of the Currency is pushing forward with stricter oversight of the 318-billion-dollar stablecoin market. Key proposed rules include limiting issuers to a single stablecoin brand, banning yield or rewards on stablecoins, requiring full one-to-one liquid reserves, fast redemption capabilities, capital minimums, and monthly executive reporting. This is essentially pushing stablecoins toward a narrow banking model. It's a framework that would benefit large, well-capitalized issuers like Circle and squeeze out smaller players.
Cross the Atlantic and the digital euro is having its own problems. Members of the European Parliament are stalling the project. The pitch for the digital euro has always been about reducing Europe's dependence on Visa and Mastercard for payment processing and strengthening financial sovereignty. But MEPs have legitimate concerns about privacy, about the impact on commercial banks, and about whether a CBDC is really the right tool for the job.
A thoughtful piece from Cornell's business school ties all of this together, noting that we're in a transition from the experimental phase of digital currencies to a structured, regulated phase. The GENIUS Act in the U.S. already mandates full reserves and audits for stablecoin issuers. Europe's MiCA framework is imposing transparency requirements. The UAE and Singapore are building interoperable digital payment systems.
From a Bitcoiner's perspective, all of this reinforces the value proposition. CBDCs are government-controlled money with surveillance built in. Regulated stablecoins are becoming bank products with bank-like restrictions. Bitcoin remains the only truly permissionless, censorship-resistant monetary network. The more the regulated world tightens its grip on digital dollars and digital euros, the more clearly Bitcoin stands apart as something fundamentally different.
Here's what I'd sit with today. Whether it's AI compressing drug development timelines from fifteen years to four, or Bitcoin's privacy layer getting quietly but steadily more sophisticated, or regulators trying to force digital money into old banking molds, the pattern is the same. The future isn't arriving with a single dramatic announcement. It's arriving through thousands of small, deliberate technical decisions made by people who understand what they're building. Pay attention to the builders, not the headlines.