# AI Drug Discovery - Week of May 28 – Jun 4, 2026: 173 Programs, Zero Approvals, the $10B Question AI Biotech Can't Yet Answer > AI drug discovery newsletter for the week of May 28 to June 4, 2026. Isomorphic Labs' $2.1B raise at a reported ~$10B valuation crystallized the bull/bear fault line, with 173 AI-originated programs in the clinic and none yet approved, while Eli Lilly's CAIO laid out a de-risking-not-speed thesis and Recursion and Schrödinger rallied without a word of podcast coverage. ## AI Drug Discovery ### Week of May 28 – Jun 4, 2026: 173 programs, zero approvals, the $10B question AI biotech can't yet answer --- ```request-access variant: ghost heading: Want this kind of synthesis on your own coverage? buttonText: Request access ``` ## TL;DR - The week's center of gravity was **Isomorphic Labs' $2.1B raise at a reported ~$10B valuation**, a pre-clinical company with no approved drug, which crystallized the bull/bear fault line in AI drug discovery. The hinge statistic: **173 AI-originated drug programs are in clinical development and none has yet been approved** [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3Duy8M_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOKWW9vHiBZRc1shaMF1N3-2BnyGkGCmPOaxXjivHeREPaVfGLpHFXfhHarSdb-2BVdW1zzIM2UDsGLWYSBpICKNeAbhAa-2F9K6v4CaUCgxsMZOFG7IQ7LTo9ThAtmJ8S1KRZLhA-3D-3D). - The real debate has moved past "does AI work" to **where in the value chain it pays off**, discovery vs. the clinic, and **whether technology or regulation is the binding constraint**. - **Eli Lilly (LLY)** was the only coverage name with both substantive AI strategy commentary (CAIO Thomas Fuchs) and a heavy week of business development, four licensing/discovery deals, closing at an all-time-high-adjacent **$1,125.38 (+4.3% on the day)**. - **Recursion (RXRX, $3.80, +9.5%)** and **Schrödinger (SDGR, $15.85, +6.2%)** rallied hard on the day but were entirely absent from the podcast conversation, worth flagging given how loud the broader AI-biotech narrative was. --- ## What's new **Isomorphic Labs dominates the tape.** On *The Heart of Healthcare*, digital-health investor Hallie Tecco walked through the DeepMind spinout's **$2.1B round**, backed by VCs and sovereign wealth funds, at a valuation she called "astronomical for a pre-clinical discovery company" (~$10B). Its named pharma partners: **Novartis, Eli Lilly, and Johnson & Johnson** [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3Do2DQ_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOM8Xd9o8tmSL2CwnrSB8-2BarWzlHF1p-2BKruiDMJ0RRlznEpv9BemFcgzvOq-2FDJ8-2FbWeYpPHPOencnx2AFGgh5rgL8N1IEtzvxi4Gz3SvccMjzDxvrtozVdP16ir4IUcm5vw-3D-3D). **The regulatory backdrop got messier.** Per the same episode, the **FDA launched a real-time clinical-trial review pilot with AstraZeneca and Amgen**, cloud-based review of safety signals and endpoints as they accrue, which the FDA's chief AI officer claimed could *cut trial duration in half*. But FDA Commissioner Marty Makary resigned (reportedly over pressure to approve flavored e-cigarettes), and co-host Steve Krause relayed that biotech investors "are just kind of almost beside themselves. They need to have some predictability in this agency" [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3Dt4zM_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOH7AJaATrZaDoXCTQDBewS7oDGSGansKPRIEegWHRFdTTiDe-2BkzXq52g494ofVgX9G5DFQV29CPGyTYX3T0QnvY52CKBwq6CXoF7-2B-2B1gpVGIbwJhicb2-2FnbDLJtUAb1RFw-3D-3D). **Frontier-AI labs keep wiring into pharma.** Tecco noted **Anthropic appointed Novartis's CEO to its board**, a signal that foundation-model companies are taking drug discovery seriously [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3Dgcjh_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOIwf0EsUUrbBWU2GH28pLGIyBch1wMGpbkS-2F-2Fq4Z1k-2F-2B2ZPQbb5uRhUWVIOcDoyp-2FRYvAHw1clG0oQ6qhqIGqL7Y3TurxjaUzmHCLitLK0lbVu0reo4VI6J96ORohlQyAw-3D-3D). Separately, on *Possible*, Reid Hoffman pitched his startup Manus AI as a "drug discovery factory for monopolies," claiming it has "already discovered novel chemistry" and replicated a decade of prior work "in hours" via a proprietary RL system he called "SIDRL" [Possible](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOjOfNBOiLZX1aJxoFw-2Byi9gFP9YJ3sdJawBvzJIF3wSr4Y7aHOjgJD2y-2FqD2yA5jt2CL3rnqR-2BZFMSj1UbI-2B14OkWjkYe9j3EObf1kP3gqI2A-3D-3DOl5I_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOCViEz-2FjK7SlnEIqsmVR41GGJ8FiYwPhgRS9Jv0ASSORjR-2FN8-2B0mpdCQDwD0uL5goWNxc8EcgvFO67zHcxUlAFLdNq1CGiTzLFZM0dOb-2Fm524k3kxaJ2RqHixl7at0kwxQ-3D-3D). (Caveat: Hoffman is a Manus co-founder, and no candidates or financials were disclosed.) --- ## The debate: does AI actually deliver ROI in pharma R&D? This is no longer a bull-vs-bear binary; it splits on three axes. **Where does AI pay off?** Eli Lilly's Chief AI Officer **Thomas Fuchs** drew the sharpest line. AI's value, he argued, is **de-risking, not speed**: "The goal is not to accelerate that dramatically because... biology has to play out. You cannot accelerate biology... What you can do is you can de-risk it by actually coming up with better molecules, safer molecules" [The AI in Business Podcast](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOgeS8gAOkODa1oxrHH8n54Rm-2BHOD2ixrgg9ercAFspEs0BfTDttdW5IilEIriJKhO58hAIYT8H1JEjcFmJgne94u7L3y77q-2BhVexTJV33MZTQ-3D-3Drqur_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOPFutBATahoawhU1ad7FBj5ZdyjOoGfYq1GqnFdJ395Uc4OdlE32SF4A6Jg1Cz8tuKxRqS4UaWZmpWmr5VPGNwcw-2B7Wu-2BJdnJ8Db11BpiHvG2eW7Z7gFM9QS2pdFi2Wv-2Fg-3D-3D). Tecco made the bear-side corollary: AI names cluster in **discovery, "not the most expensive phase"**, the costly work is trials, cohort selection, and safety/efficacy, "where biology still does what biology does", leaving a capital gap: "who's going to fund the expensive piece?" [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3DxyL8_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOPNfiGbiXT7mLsGSN28PsilVLsesH6cjbEn40QPQX08l3jcxJLt4aqXM3gD64NmXCj7FWpffkwFbunHrEyHaSAwpJCxElCFqpzmCnW4INL16I6V1tJkIyHxhvqMxPmYNVg-3D-3D). **Tech or regulation as the binding constraint?** Logos LP's **Peter Mantas** delivered the week's most contrarian framing on *This Week in Intelligent Investing*. He is wildly bullish on the input, "AI just brings the cost of knowledge to zero... the number one beneficiary [is] biotech", predicting "the biggest biotech bubble we've ever seen, make 2021 look like child's play." Yet he insists the bottleneck is not technology: "The only way you break Eroom's Law is changes in regulation. Not technological disruption... if our regulators and policymakers don't address that problem, then AI, all it does is just create a queue" [This Week in Intelligent Investing](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOjvApT-2Ba4RHRiL1keHjrY-2BuazGLXb8Kohj64Yq-2BGTyNWWYTi4K-2FI5ahsY3c3kidlW9UVU8juEXcjVCW5UHAcZ69tGd-2FVhj6fFSqEc-2FPffUTlA-3D-3DFXCI_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowONaFjFXNbzpj5huER7-2BlEGHC80P5WWw3QW7l2J3Vrj1nLWF-2FJpLNPyVKHrYqRVOThA7H8rcww-2Boi4TTMmE23Zyu9hEg-2Bl1aKbx1sxtMW47REtv9IUXpImRpyTL-2BJVJB2JQ-3D-3D). **Is the hype redeemable?** The blunt skeptic was **Peyton Greenside, CEO of Big Hat Bioscience**, quoted via Stat News: "AI in drug development is powerful, but it's overhyped", she doubts AI compresses the *full* development timeline the way the narrative claims [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3Dpupf_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowONsq7ofBNUArWN3CX9OtGWGchRn0-2Ba4A6QgMTyexAzrTDF209f4adHujDYiL1jVfhgaybtuNkeslz-2FYpAIlcEtNvjIroKENhpWmtiEjdVkEoXCd7-2FCVLjpydyHqxcx037g-3D-3D). Against that, Hoffman leans on TAM: drugs are "essentially legal 20-year-long monopolies," and he uses GLP-1s, multiple players each generating tens of billions, as proof the category isn't winner-take-all [Possible](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOjOfNBOiLZX1aJxoFw-2Byi9gFP9YJ3sdJawBvzJIF3wSr4Y7aHOjgJD2y-2FqD2yA5jt2CL3rnqR-2BZFMSj1UbI-2B14OkWjkYe9j3EObf1kP3gqI2A-3D-3DaEbY_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOEW5kQgMIIvnUggLwJvIUm3klIfwysQXCL5Dw0QQCIrCgokB7-2B-2Fc9fWCPVnG5vYqQ7On0brcvRfJeUkO7J-2BixokOsDDNr-2BVQtLovcq05K9qIqH6cTD7xz-2F5N3iKyYg-2Bpvg-3D-3D). One quieter but consequential disagreement: Fuchs's **architectural** view that the productive engines are "molecular models, diffusion models, generative flow models", not LLMs, which Lilly uses "mostly just to orchestrate work", cuts against the "foundation models will solve biology" framing implicit in much of the bull case. "Language models will never be able to be actually a good scientist," he said [The AI in Business Podcast](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOgeS8gAOkODa1oxrHH8n54Rm-2BHOD2ixrgg9ercAFspEs0BfTDttdW5IilEIriJKhO58hAIYT8H1JEjcFmJgne94u7L3y77q-2BhVexTJV33MZTQ-3D-3DjdF2_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowODDGjpL0y06AZtTMXHRvs7yuIPNWk4Jo-2BgY4lH4BDsq8ELuCRGTU9qXyai-2F6idozKoDQqv1lrXciyKKNglz0rnG-2B1wvgr-2FrDefEyiepD3gBJDcXl9aeN7wK00C5Kl15f8Q-3D-3D). ```request-access variant: inline heading: Want this kind of synthesis on your own coverage? buttonText: Request access ``` ## Stocks in play **Eli Lilly (LLY), $1,125.38, +4.3%, ~$1.06T mkt cap.** The clearest read on how an incumbent operationalizes AI. Fuchs disclosed Lilly is deploying an **NVIDIA DGX SuperPOD B300 with 1,000 B300 GPUs**, "the most powerful supercomputer in our industry", and framed Lilly's decades of unpublished *negative* experimental data as a structural moat: "we know a lot of things that do not work, and that helps us... build better molecules" [The AI in Business Podcast](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOgeS8gAOkODa1oxrHH8n54Rm-2BHOD2ixrgg9ercAFspEs0BfTDttdW5IilEIriJKhO58hAIYT8H1JEjcFmJgne94u7L3y77q-2BhVexTJV33MZTQ-3D-3DOg8N_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOEv6myCb4j9SawHlKNLaF1k7zVIHWyMMmn-2FxWoKxJOd1YZeiNvL97SPWDAX1rvC0Dn2Mn-2FO8DsT-2FxXLpf-2BJ7cfaHq2tzxjdtUotFeItdU099SakNkK3cML4A-2F6YzwBoDug-3D-3D). Its **TuneLab** hub hosts validated internal and partner models. The fundamental tape was busy too: Lilly's Retevno hit its primary endpoint in a Phase 3 NSCLC trial (83% reduction in recurrence/death vs. placebo) and the company signed four BD deals this week, including a discovery collaboration with Haisco worth up to **$3.05B** and an RNA-editing deal with Ascidian worth up to **$1.9B**. **Recursion (RXRX), $3.80, +9.5%, $1.7B mkt cap.** Shares popped on the day but trade near the low end of a $2.77–$7.18 52-week range, and the company is loss-making (trailing EPS -$1.16). Notably, RXRX drew **zero podcast commentary** this week even as Isomorphic's raise dominated the AI-biotech conversation, a narrative-share gap to monitor. **Schrödinger (SDGR), $15.85, +6.2%, $1.2B mkt cap.** Also rallied (off a $10.95–$27.63 range, trailing EPS -$1.40) with no podcast coverage. The physics-based platform story simply wasn't part of the week's discourse. ```request-access variant: inline heading: Want this kind of synthesis on your own coverage? buttonText: Request access ``` ## Read-throughs - **Isomorphic's ~$10B mark is the comp that pressures the publics.** If a pre-clinical platform with no approval commands ~$10B private, it sharpens, rather than supports, scrutiny of public AI-biotech multiples; the bear read (Tecco, Greenside) is that discovery-stage value is being capitalized before the expensive clinical proof exists [The Heart of Healthcare](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOimV5mHZ9wHvHicS-2Bz3kjvb0jDSGcK3LSdeboWw1vfyoTYt8IxwOMBV38sDBYpXT0y1qI2iQb-2Flx3fZq-2BivRQaA7H-2Fc3-2F4mByS9-2Bp-2BgXpv8dw-3D-3DISfV_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOMP7mgiJOfyLyiZGcpdO2qSjul2TfDIPRdmCB3zdUu3RG11zreA5MCB7g5Rye-2FXS5rlPy7giw6jj7ITwLaaST-2FwWFaF3tiRXj4rVtmUyJZ1k0-2FE7fn7ck8Qqz-2Fkqjm-2BqVg-3D-3D). - **LLY is a partner, not a pure-play, and that's the point.** Lilly sits inside Isomorphic's partner roster *and* runs its own B300/TuneLab stack, capturing AI optionality while its GLP-1 franchise funds the expensive clinical phases others can't [The AI in Business Podcast](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOgeS8gAOkODa1oxrHH8n54Rm-2BHOD2ixrgg9ercAFspEs0BfTDttdW5IilEIriJKhO58hAIYT8H1JEjcFmJgne94u7L3y77q-2BhVexTJV33MZTQ-3D-3DhTYU_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOJt6shrj9lrqQjy41-2FsiLgkh0JlvcRqgmf5I4AbadQW7I49LfgSAq0AS8xXQbVKn7K08kkVo6tp7Ue6Rfs07Pl-2FUou2WsZgJQr4-2BMMSzW5E7G72Bv5kyvpXSIP76Lt8gqg-3D-3D). Hoffman's own GLP-1 market-structure argument inadvertently underwrites Lilly's incumbency [Possible](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOjOfNBOiLZX1aJxoFw-2Byi9gFP9YJ3sdJawBvzJIF3wSr4Y7aHOjgJD2y-2FqD2yA5jt2CL3rnqR-2BZFMSj1UbI-2B14OkWjkYe9j3EObf1kP3gqI2A-3D-3DBw9s_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOPyWqmRoLprI0ddT8n2Xp7jkPYQBsCirgBsRJAzmBtY5wS2bPLU9-2FYoadrSufzY1zGWM9zVbIxapvUgjHVpAtojyGMLPuwRCP7VfKpjebnvndK7urN6obp7zafmqowtFRA-3D-3D). - **Regulation, not models, may set the clock.** Mantas's "AI just creates a queue" thesis plus the Makary resignation argue that FDA throughput and predictability, not algorithmic progress, gate near-term value capture. The AstraZeneca/Amgen real-time-review pilot is the offsetting positive to watch [This Week in Intelligent Investing](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOjvApT-2Ba4RHRiL1keHjrY-2BuazGLXb8Kohj64Yq-2BGTyNWWYTi4K-2FI5ahsY3c3kidlW9UVU8juEXcjVCW5UHAcZ69tGd-2FVhj6fFSqEc-2FPffUTlA-3D-3DBFB__7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOLsPSRXU0IH574ZDeDMtdUGa5VdJPo8mb6iCiNKAoL3rVZ5pGUlKzDdHl0euPmm0TF93TiNYEMbrkJBxKFgel9b8zpXRamKnniaoQuweMpULeoSGQ9gQoReXwzBCxHbO6Q-3D-3D). - **"De-risking, not acceleration" reframes the KPI.** If Fuchs is right, the metric that matters isn't shorter timelines but higher Phase 2/3 success rates, a slower-burn, harder-to-see catalyst than the market's "compress a decade to two years" narrative [The AI in Business Podcast](http://url7324.matterfact.com/ls/click?upn=u001.idHmPrr2Geh7KYLAsTy7NkrIVb-2FgA4pmf2rMXQwGcOgeS8gAOkODa1oxrHH8n54Rm-2BHOD2ixrgg9ercAFspEs0BfTDttdW5IilEIriJKhO58hAIYT8H1JEjcFmJgne94u7L3y77q-2BhVexTJV33MZTQ-3D-3DuXfr_7mLGwmUci-2BLaXswv9WX1yTgqn3Wad-2FotHhzHgSNAZbX-2FKUhZskmCG0wIEuQSsjVQbp7ghkLlJeG8O0Y9udowOAfF-2BD3Ljek0tjIebF6wOhr5jffP02ZOjyIk-2BpsyaCNj2ycKianYUS6B6p03gU5K7WgMz6OJfLmz0Scw1dcEUSdawNu5bF81HxjBz682N02RRCxteGWgtYZIa3MhaD1WYw-3D-3D). ## What changed vs last week This is the **inaugural issue** of AI Drug Discovery Weekly, so there is no prior baseline to compare against. Going forward this section will track week-over-week shifts in narrative, valuation, and the bull/bear balance. Two markers to carry into next week: (1) whether the 173-programs / zero-approvals statistic moves, and (2) whether Recursion or Schrödinger re-enter the podcast conversation after a week of silence despite sharp share-price moves. ```request-access variant: banner heading: Run your own corpus. description: matterfact is deployed with select institutional partners. Request access to run it on your own coverage. buttonText: Request access ```