Anthropic CFO Krishna Rao talks about how his team uses Claude. Says Head of Tax is biggest power user working on tax policy engine. Finance team built 70 specific skills, including automating financial statements and monthly reviews (says they are 90% done, then human reviews).

The MVP of the Trump-Xi summit so far has been Fox’s Bret Baier. Walking through wild traffic to frame a shot. Ordering a sausage from a humanoid robot. Reporting on how the Fox crew got ticketed $40 for illegally parking after getting spotted by CCTV camera.

“mom, how did we get so rich?” “your dad went on prediction markets and bet our whole retirement portfolio on the Trump and Xi handshake lasting more than 10 seconds”

NEW ON THE POD: I tried something different this week. Instead of bringing on guests to interview, I pulled in two co-hosts — Teira Gunlock, CEO of First Stop Health, and Dr. Eric Bricker, CMO of AHealthcareZ — and we just… rumbled. Three topics, no script, healthcare's messiest debates. • AI regulation and whether EMRs need to be forced open (Eric made a compelling Steve Jobs / App Store analogy I can't stop thinking about). • First Stop Health hitting 5x productivity on their first fully AI-native dev cycle. • ICHRA vs. captives and why small/mid employers keep getting crushed on renewals. • Laser exclusions on stop-loss, the 80% renewal cycle, and why ICHRA's network problem is the real adoption blocker.  • Whether any of this is actually good for patients vs. just good for employers, the federal government, and everyone else trying to pass the healthcare cost hot potato (the “not it” dynamic). Also: Eric lost a chicken the day we were supposed to meet for lunch. So there's that. Apple: Spotify: 𝐓𝐈𝐌𝐄𝐒𝐓𝐀𝐌𝐏𝐒 00:00 — Co-hosts, not guests: trying a new format 01:05 — Tiera introduces First Stop Health (and her philosophy degree, which becomes relevant) 01:49 — Eric's origin story: Hopkins, Compass, YouTube, and accidentally becoming healthcare's finance explainer 07:23 — AI regulation: iterative governance vs. exponential tech (spoiler: they don't meet well) 10:48 — The liability question and why physicians are still on the hook for every AI decision 12:30 — Eric's App Store argument: why EMRs need to be forced open for agentic AI to matter 16:01 — "It's clinical workflow and billing software" — Tiera's correction that reframes the whole EMR conversation 17:11 — Is it malpractice NOT to use AI? (yes, eventually) 18:58 — First Stop Health's 5x dev cycle productivity gain (and what it means for clinical workflows) 22:44 — Radiology, AI, and the first real physician-specialty price war 28:09 — How to actually change the incentives (Tiera answers, Eric hides) 30:15 — The canary in the coal mine: HCA, Tenet, UHS will be first to layoff behind AI 32:05 — The "reskilling vs. swap-out" moral imperative (— free idea for every health system CHRO) 35:03 — AI primary care bots for rural patients: closing inequities or compounding them? 41:22 — Eric's "emotional labor is the next frontier" thesis (the best line of the pod) 46:04 — ICHRA explained: defined contribution, the 401k analogy, the Sally-will-quit problem 51:30 — Group captives: collective bargaining for stop-loss coverage 56:44 — Why exchange networks kill ICHRA adoption in the mid-market 01:00:47 — What IS the role of the employer in healthcare in 2026? 01:07:38 — Uwe Reinhardt, Donna Shalala, and why campaign finance reform is healthcare reform 01:12:51 — Keeping it in perspective: 1968 was worse. Also, Texas won the title in '69.

Sanjay Mehrotra legendary LinkedIn work experience: > co-invents tech for NAND flash memory > co-founds SanDisk and runs for 28 years > sells SanDisk to Western Digital for $19B in 2016 > becomes Micron Technology in CEO in 2017 and market cap since grown 29x from $30B to $860B

Still incredible that the DeepMind documentary has footage of exact moment Demis is told that AlphaFold can “easily” predict all known (1-2B) protein sequences “in a month” and he says to do it. Then, it shows the moment AlphaFold is released to the world.

Masa took Arm private for $32B in 2016. Went pubic again in 2024 and market cap is now $234B. He owns 87%, which is around $204B. So, Masa is now up $172B on Arm, which is more than 10x the amount he lost on WeWork ($16B). Wild.

Australian tech entrepreneur Paul Conyngham explains how he used ChatGPT/AlphaFold (spent $3,000 with no biology background) to create a custom MRNA vaccine to treat his dog’s cancer tumors. Unreal.