SIGNAL//SYNTH
Ai

Hippocratic AI’s Munjal Shah on Building Safe AI

aired Nov 18, 2026
Signal
84.6/ 100
High signal
confidence 0.99
Orig53.4
Actn100.0
Dens100.0
Dpth100.0
Clty79.8
Summary

Like any good scientist, we're just empiricists. And so, you know, one of the things that we did differently than everybody else, which, you know, lately the, the debate about AI safety and I need the government to make, to help me make it safe, that certain people have been spouting, you know, which drives me nuts because we took a very different approach.

Why listen

It goes beyond the title with direct discussion of like, know, yeah, including: We just kept testing it ourselves.

Key takeaways
  1. 01We just kept testing it ourselves
  2. 02We didn't have confidence until we built it and tested it and saw
  3. 03They're like, "When are you shipping?" And I'm like, "When it's safer than a human." And they're like, "No, no, no, really, when are you shipping?" I'm like, "No, really." Like, "D
Best for
listeners looking for a practical AI episode debrief