Get the app
extract 0 just added 0

aitrainingjobs/red-teaming-ai-safety-market

published by @aitrainingjobs

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
26
Last 30 days
26
Likes
-
Size
474 KB
Created on HF
2026-10-19
Age
2 days ago

Description

The Market for Red-Teaming & AI Safety Work

AITraining.jobs Research Report, data from 20 Sep 2026 to 17 Oct 2026. By Yuma Heymans, AITraining.jobs (O-mega Enterprise, Inc.), Newark, Delaware, United States.

Summary

On 17 Oct 2026, 54 roles in Red-Teaming & AI Safety were open on 12 platforms (Alignerr listed 33.3%); their median advertised rate was $70/hr, 8% above the $65/hr of all AI training work; each open role drew 1.13x the apply clicks of the average role.… See the full description on the dataset page: https://huggingface.co/datasets/aitrainingjobs/red-teaming-ai-safety-market.

Tags

language:enlicense:cc-by-4.0size_categories:n<1Kformat:csvmodality:documentmodality:tabularmodality:textlibrary:datasetslibrary:pandaslibrary:polarslibrary:mlcroissantregion:uslabor-marketai-training-datahuman-datajob-postingsrlhf

README current version from Hugging Face

license: cc-by-4.0
language:

  • en
    pretty_name: "The Market for Red-Teaming & AI Safety Work"
    size_categories:
  • n<1K
    tags:
  • labor-market
  • ai-training-data
  • human-data
  • job-postings
  • rlhf
    configs:
  • config_name: work-open-roles-weekly
    data_files: work-open-roles-weekly.csv
  • config_name: work-listing-flow
    data_files: work-listing-flow.csv
  • config_name: work-platforms
    data_files: work-platforms.csv
  • config_name: work-pay
    data_files: work-pay.csv
  • config_name: work-pay-bands
    data_files: work-pay-bands.csv
  • config_name: work-demand
    data_files: work-demand.csv
  • config_name: work-demand-by-age
    data_files: work-demand-by-age.csv
  • config_name: work-location-rules
    data_files: work-location-rules.csv
  • config_name: work-countries
    data_files: work-countries.csv
  • config_name: work-asked-for
    data_files: work-asked-for.csv
  • config_name: work-requirements
    data_files: work-requirements.csv

The Market for Red-Teaming & AI Safety Work

AITraining.jobs Research Report, data from 20 Sep 2026 to 17 Oct 2026. By Yuma Heymans, AITraining.jobs (O-mega Enterprise, Inc.), Newark, Delaware, United States.

Summary

On 17 Oct 2026, 54 roles in Red-Teaming & AI Safety were open on 12 platforms (Alignerr listed 33.3%); their median advertised rate was $70/hr, 8% above the $65/hr of all AI training work; each open role drew 1.13x the apply clicks of the average role.

Question: Who hires for Red-Teaming & AI Safety work, what does it pay, how much competition is there for each opening and who may apply?

Tables

  • work-open-roles-weekly.csv: Open roles, week by week
  • work-listing-flow.csv: Listings added and removed
  • work-platforms.csv: Who lists it
  • work-pay.csv: What it pays
  • work-pay-bands.csv: How its rates are distributed
  • work-demand.csv: Applicant demand per open role
  • work-demand-by-age.csv: Applicant demand by listing age
  • work-location-rules.csv: Who may apply: location rules
  • work-countries.csv: Who may apply: roles open to each country
  • work-asked-for.csv: What the roles are called
  • work-requirements.csv: What the descriptions ask for

Paper

The full paper with method, limitations and disclosure is paper.pdf in this repository.

Source and method

The public job boards of the AI training platforms this site tracks, read every day. Apply clicks are made on AITraining.jobs; full descriptions come from the platforms whose boards publish them.

The live, daily-updated version of this report is at https://aitraining.jobs/research/red-teaming-ai-safety-market?utm_source=huggingface&utm_medium=repository&utm_campaign=research-2026-10.

License and citation

CC BY 4.0. Cite as:

Heymans, Y. (2026). The Market for Red-Teaming & AI Safety Work. AITraining.jobs Research Report, data from 20 Sep 2026 to 17 Oct 2026. https://aitraining.jobs/research/red-teaming-ai-safety-market

Disclosure: AITraining.jobs receives referral fees from some platforms it lists; this does not change any figure.

README history 2 revisions

Every night we snapshot the README of every dataset in the catalog. When the SHA changes we archive the new version and diff it against the last. This is the evolving thought record of the alignment-data field: what the author decided to say about the corpus, and how that framing shifted over time.

  1. 2026-10-19The Market for Red-Teaming & AI Safety Work (version 1, 18 Oct 2026)7a9b56f2.9 KB
    Loading...
  2. 2026-10-19initial commita1522110 B
    Loading...

Used by abliterated models none yet

No abliterated models in our catalog have declared this dataset in their YAML metadata yet. This can mean the dataset is used but not declared, is used outside abliteration workflows, or is new. The link will populate automatically as models are indexed.

Related datasets same stage · extract

Anthropic/hh-rlhf
2.1M dl 2442 likes 90.4 MB
Dataset Card for HH-RLHF Dataset Summary This repository provides access to two different kinds of data: Human ...
djain95/sae-jailbreaks-cache
468.6K dl 1 likes 30.6 GB
mosaicml/dolly_hhrlhf
309.5K dl 113 likes 23.7 MB
Dataset Card for "dolly_hhrlhf" This dataset is a combination of Databrick's dolly-15k dataset and a filtered subset of Anthrop...
Dahoas/full-hh-rlhf
266.3K dl 104 likes 129.8 MB
Dataset Card for "full-hh-rlhf" Anthropic's HH dataset reformatted into prompt, chosen, rejected samples.
mlabonne/harmful_behaviors
198.2K dl 169 likes 23 KB
mlabonne/harmless_alpaca
190.2K dl 51 likes 1.2 MB
walledai/AdvBench
176.3K dl 123 likes 38 KB
Dataset Card for AdvBench Paper: Universal and Transferable Adversarial Attacks on Aligned Language Models Data: AdvBench Datas...
JailbreakV-28K/JailBreakV-28k
116.4K dl 80 likes 288.8 MB
⛓‍💥 JailBreakV-28K: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks 🌐 ...

Read further

Community best practices (extraction)
5 rules the field converged on for building the pair.
Building your own extraction pair
When the reference pair is not enough. Sizing, framing, and evaluation.
Datasets that shaped abliteration
The five corpora that did most of the work.
Non-English abliteration communities
How Chinese, Russian, Japanese, and Korean practitioners build extraction pairs.

Metadata

License
cc-by-4.0

Download the dataset

0

Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.

Recommended: datasets library (Python) 
from datasets import load_dataset
ds = load_dataset("aitrainingjobs/red-teaming-ai-safety-market")
Raw snapshot (Python) 
from huggingface_hub import snapshot_download
snapshot_download(repo_id="aitrainingjobs/red-teaming-ai-safety-market", repo_type="dataset")
git clone (requires git-lfs) 
git clone https://huggingface.co/datasets/aitrainingjobs/red-teaming-ai-safety-market
Browse repository on Hugging Face ->
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration