Benchmark Almanac

ETHICS

ALBERT-xxlarge 71% tops its leaderboard
Not in model cards

Models predict common moral judgments about text scenarios covering justice, well-being, duties, virtues and commonsense morality.

ICLR 2021. Has normal and hard test splits.

Released
August 2020
Kind
Deception and alignment
Level
Everyday
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public

In the authors’ words

We show how to assess a language model's knowledge of basic concepts of morality. We introduce the ETHICS dataset, a new benchmark that spans concepts in justice, well-being, duties, virtues, and commonsense morality. Models predict widespread moral judgments about diverse text scenarios. This requires connecting physical and social world knowledge to value judgements, a capability that may enable us to steer chatbot outputs or eventually regularize open-ended reinforcement learning agents. With the ETHICS dataset, we find that current language models have a promising but incomplete ability to predict basic human ethical judgements. Our work shows that progress can be made on machine ethics today, and it provides a steppingstone toward AI that is aligned with human values.

Abstract of Aligning AI With Shared Human Values (arXiv:2008.02275, CC0)
6
Entries on its own leaderboard, 0 models we track

ALBERT-xxlarge tops the leaderboard with 71%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

  1. 71.0 ALBERT-xxlarge
  2. 68.0 RoBERTa-large
  3. 56.1 BERT-large
  4. 51.6 BERT-base
  5. 39.3 GPT-3 (few-shot)
  6. 33.5 Word Averaging
  1. 1 ALBERT-xxlarge standard 71%
  2. 2 RoBERTa-large standard 68%
  3. 3 BERT-large standard 56.1%
  4. 4 BERT-base standard 51.6%
  5. 5 GPT-3 (few-shot) standard 39.3%
  6. 6 Word Averaging standard 33.5%

Source: Official ETHICS leaderboard

#EntryScoreSettingDateSource
1 ALBERT-xxlarge 71% standard Official leaderboard
2 RoBERTa-large 68% standard Official leaderboard
3 BERT-large 56.1% standard Official leaderboard
4 BERT-base 51.6% standard Official leaderboard
5 GPT-3 (few-shot) 39.3% standard Official leaderboard
6 Word Averaging 33.5% standard Official leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Safety and alignment

All 465 in Safety and alignment
  • FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
  • StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is. Inkling 98.6%
  • Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
  • DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics. Muse Spark 1.2 88.4%
  • Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
  • IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%