Listen now | Eric Nguyen explains why biology needs specialized AI, how DNA can be modeled as language, and why lab verification remains a key bottleneck between AI predictions and discovery. ✦ AI generated
The challenge of safety in biological AI is not about removing capabilities but about preserving usefulness while reducing risk.
Eric Nguyen: How to make powerful biology models safer without eliminating the capabilities that make them valuable
Radical Numerics is building multimodal models that bridge across biological modalities instead of tackling each in isolation.
Eric Nguyen: Radical Numerics is now building models that can connect information across DNA, RNA, proteins, epigenetics, and other parts of biology, rather than treating each as a separate problem.
Eric cites concrete results: Evo generated viable bacteriophage genomes, and Omnii replicated two years of Alzheimer's lab findings in days.
Eric Nguyen: Researchers have already used Evo to generate viable bacteriophage genomes, and Eric says Radical Numerics' newer model, Omnii, matched key findings from two years of Alzheimer's wet-lab research in a matter of days.
One of the key bottlenecks between AI predictions and biological discovery remains the speed and difficulty of wet-lab verification.
Eric Nguyen: Why testing AI-generated biology in the lab is still slow and difficult
Eric Nguyen opens the episode by framing the core tension: AI's ability to generate and manipulate life offers transformative potential for biology, but also introduces serious responsibility and risk.
Eric Nguyen: The potential to create or manipulate life with AI could reinvent nearly all of biology, but it also carries inherent risk. We've seen just a taste of this on the natural-language and chatbot side. Giving AI the power to generate and create life carries a certain level of responsibility.
Eric acknowledges the dual-use risk of generative biology AI and explains that Radical Numerics is investing in both design capabilities and defense against misuse.
Eric Nguyen: He also believes these tools could make it easier to create dangerous pathogens, which is why the company is working on both biological design and biodefense.