# AI Alignment & Safety: State of the Field (October 2026)

Research dossier · v1 · updated 2026-10-30 · 5,488 words · 122 sources · 30 claims · verification: verified

An expert briefing on AI alignment and safety as of late October 2026: preference optimization (RLHF, DPO, AI feedback, verifiable rewards), reward hacking, deception and scheming evaluations, chain-of-thought monitoring, AI control, scalable oversight, misuse safeguards, dangerous-capability evaluation and the governance built on them. The main takeaway is that post-training reliably shapes everyday behaviour but verification is getting harder: reward hacking learned in production-style training can generalize into broader misalignment, frontier models increasingly recognize when they are being tested, and developers now report a model crossing their highest cybersecurity risk threshold. Safety claims therefore rest more on monitoring, access controls and layered safeguards than on direct evidence that a model's goals are aligned.

## Contents

- Executive summary
- Background and core concepts
- Timeline of key milestones
- State of the art
  - Preference optimization in practice
  - Verifiable rewards and specification-based training
  - Reward hacking and overoptimization
  - Deception, scheming and alignment faking
  - Monitoring, auditing and AI control
  - Scalable oversight and automated alignment research
  - Misuse resistance
  - Dangerous-capability evaluation and deployment decisions
  - Quantitative benchmarks
- Open problems and active debates
- Major programs and players
- Outlook: what to watch (November 2026 to April 2028)
- Key papers
- Glossary

## Sample findings

- Labelers preferred outputs from a 1.3B-parameter InstructGPT model, trained with RLHF, over those of the 175B-parameter GPT-3.
- AI-generated feedback gave policies performance comparable to RLHF on summarization and on helpful and harmless dialogue.

## Get the full dossier

GET https://space-kitty.com/api/v1/dossiers/alignment-safety costs $0.20 via x402 (USDC on Base Sepolia (testnet)). Send `Accept: text/markdown` for Markdown or `application/json` for structured data (claims with citations, timeline, benchmarks, open problems, glossary, bibliography).
---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.