DV
P
AI Safety Lab

AI Safety Lab

60 days · 12 weeks · 3 parts · From informed to expert. Part A builds the applied craft — threat modeling, safety taxonomies, red-teaming, and evaluation. Part B makes you literate in alignment research — the alignment problem, deceptive alignment, interpretability, and scalable oversight. Part C is governance — risk frameworks, regulation, and a capstone safety-evaluation program. Theory + hands-on artifacts + curated, verified primary sources.

→ Start here: how this track works → Portfolio: the binder you'll walk out with

0 / 60
Days Complete
0 / 12
Weeks Complete
0%
Overall Progress
Start with Day 1 →
Continue
Part A — Applied Safety & Evaluations   Weeks 1–5
Part B — Alignment Research Literacy   Weeks 6–9
Part C — Governance, Policy & Systemic Safety   Weeks 10–12