Skip to content
English

Switch theme

Discover▾

Not sure where to start? This week’s hottest software, models and stacks in three minutes.

AI Apps▾

AI software people use: from assistants like ChatGPT to tools for coding, video and workflows, picked by purpose.

For Agents▾

Software for AI: tools that let your agent (Claude Code, Codex, Cursor…) browse, read docs, run code and remember.

Models▾

Choose models, estimate cost, self-host: prices, rankings, benchmarks, free tiers and compute.

Solutions▾

A goal or an error message? Ready-made stacks, steps and fixes by scenario.

Learn▾

A systematic course from zero to the frontier, with an algorithm library, must-read papers and tutorials.

Frontier▾

Daily AI news, weekly highlights, the latest papers and how AI is changing every discipline.

// Stage 9 · Safety, Evaluation & Frontier

Frontier Risk, Scalable Oversight & Governance

Weak-to-strong generalisation, debate, Constitutional AI, risk frameworks and regulation (EU AI Act, NIST AI RMF).

What you will learn ↓ Code & exercises ↓

about 6 hours4 algorithms3 paperslesson text in progress

01Goals

Do these first: Alignment: RLHF, DPO & Constitutional AI

02Algorithms to master

Each card has the original paper, prerequisites and resources.

Weak-to-Strong Generalization Supervising strong models with weak-model labels as an analogy for superalignment. 2023core AI Safety via Debate Two AIs debate and a human judges, aiding oversight of hard questions. 2018ext Constitutional AI / RLAIF Uses a written constitution so AI critiques itself; AI feedback replaces human labels. 2022core Sleeper Agents (Backdoors that Persist) Backdoored models keep hidden behaviors even after safety training. 2024ext

03Must-read papers

04Courses & videos

05Hands-on: project, code & exercises

Free, official or author-published pages. Most run in the browser or on a free Colab GPU.

PROJECT

Draft a “generative AI use policy” for a 50-person company covering data, approvals, audit and incident response.

  1. 1
    bluedot.org ↗Courses

    Join a BlueDot AI safety or governance course with group discussions.

  2. 2
    inspect.aisi.org.uk ↗Official docs

    Write a small Inspect eval for a dangerous capability or deception.

06Self-test

Answer in your head, then open to check.

What is weak-to-strong generalisation an analogy for?

Weak models supervising strong ones, like humans supervising superhuman AI.

How does the EU AI Act tier its rules?

By risk level.

The four core functions of the NIST AI RMF?

Govern, map, measure, manage.

On these routes: Product Managers & Leaders

Data updated: AI algorithms & curriculum(UTC+8)