TBPN timeline: CRISPR saves a baby, worm-inspired robots, Google PM culture, and public.com backtests venture portfolios vs S&P
May 16, 2025
Key Points
- Researchers delivered the first clinical proof that personalized CRISPR gene therapy works in living patients, treating a baby with a rare liver disorder in under six months with three doses costing under $5 million and no transplant needed.
- MIT's CSAIL lab is building soft, worm-inspired robots that can be ingested for non-invasive internal surgery, powered by AI models based on worm neural networks.
- Public.com's backtesting tool shows Founders Fund-backed IPOs returned 659% versus 150% for the S&P 500, while a synthetic portfolio of highly paid CEOs returned 2,000%.
Summary
A CRISPR breakthrough is dominating the week's science news. Researchers used personalized gene therapy to treat a baby with CPS1, a rare lethal liver disorder. Three lipid nanoparticle doses of CRISPR delivered the treatment in less than six months without requiring transplant or viral vectors. The therapy costs under $5 million today and applies to single-gene mutations in organs that can be safely reached, affecting thousands of babies annually. Jennifer Doudna, one of CRISPR's pioneers, won the Nobel Prize and is the subject of a Walter Isaacson biography. The treatment marks the first real-world clinical proof that personalized gene therapy can work in living patients.
MIT's CSAIL lab is designing soft, flexible, and edible robots inspired by worms and sea turtles. Edible robots could be ingested for non-invasive internal surgery, crawling through the body to repair internal bleeding or perform other repairs. The lab also uses AI models based on worm neural networks to power robot brains. This work represents a broader shift away from bipedal humanoid designs toward diverse morphologies.
Google's product culture remains a friction point. A former Android Auto design lead posted that 40% of their time went to arguing with problematic product managers, 20% to coaching designers, and 40% to Google's organizational overhead. Ideas discussed internally in 2015 took until 2020 to ship. The broader critique is that Google's PMs aren't fundamentally bad but that organizational structure and speed prevent products from being positioned or marketed correctly.
Public.com's backtesting feature lets users create synthetic portfolios and test them against historical market performance. Founders Fund-backed companies that went public returned 659% versus 150% for the S&P 500. A16Z returned 589%, Sequoia 123% (underperforming the S&P at 147%), and A6Z 466%. F1 team sponsors returned 317%. The standout result came from a portfolio of highly paid CEOs, which returned 2,000% versus 176% for the S&P. Bald CEOs also outperformed at 433% versus 150%, with Jeff Bezos and Brian Armstrong among the names cited. Steve Ballmer rounds out the list. Public.com presents these backtests as entertainment rather than investment guidance.