Explore how AI models Evo 1 and Evo 2 are designing functional bacteriophages from scratch, creating a new biosecurity framework for synthetic biology.

We have reached a point where artificial intelligence isn't just helping us organize biological data; it is actually generating entirely new, self-replicating viruses that have never existed in nature.
Teach me a biosecurity risk-assessment framework I can reuse when AI starts generating real-world designs. Use Stanford’s AI-designed self-replicating viruses in E. coli as the case study, and give me 3 warning signals and decision rules.




![[2510.27629v2] Best Practices for Biorisk Evaluations on Open-Weight Bio-Foundation Models](https://d1y2du6z1jfm9e.cloudfront.net/assets/podcast/blue.png)

Evo 1 and Evo 2 are advanced AI models used by researchers like Brian Hie and Aditi Merchant to design functional bacteriophages from scratch. These models are trained on the vast language of genomes across the tree of life, allowing them to generate digital architectures for self-replicating viruses. Recently highlighted in the journal Science, these models produced 700,000 potential designs, leading to the creation of sixteen living viruses that successfully infected E. coli bacteria.
AI contributes by acting as a digital architect that can write the genetic code for entirely new organisms. Because biology is unforgiving and a single misplaced nucleotide can make an organism non-viable, researchers use predictive models to ensure genetic sequences are functional. This technology allows scientists to move beyond organizing biological data to actually generating self-replicating viruses that have never existed in nature, significantly lowering the barrier to entry for synthetic biology.
A biosecurity framework is essential because AI now has the capability to generate functional, self-replicating viruses from digital blueprints. As demonstrated by the successful creation of viruses that destroy E. coli, the ability to design life through code presents unprecedented challenges. Establishing a framework helps manage the risks associated with synthetic biology and the powerful predictive models like Evo 1 and Evo 2 that are writing the future of life in research labs.
コロンビア大学卒業生がサンフランシスコで開発
"Instead of endless scrolling, I just hit play on BeFreed. It saves me so much time."
"I never knew where to start with nonfiction—BeFreed’s book lists turned into podcasts gave me a clear path."
"Perfect balance between learning and entertainment. Finished ‘Thinking, Fast and Slow’ on my commute this week."
"Crazy how much I learned while walking the dog. BeFreed = small habits → big gains."
"Reading used to feel like a chore. Now it’s just part of my lifestyle."
"Feels effortless compared to reading. I’ve finished 6 books this month already."
"BeFreed turned my guilty doomscrolling into something that feels productive and inspiring."
"BeFreed turned my commute into learning time. 20-min podcasts are perfect for finishing books I never had time for."
"BeFreed replaced my podcast queue. Imagine Spotify for books — that’s it. 🙌"
"It is great for me to learn something from the book without reading it."
"The themed book list podcasts help me connect ideas across authors—like a guided audio journey."
"Makes me feel smarter every time before going to work"
コロンビア大学卒業生がサンフランシスコで開発
