KGC-Safety (LG)
K-EXAONE 2.0 99.8% leads since July 2026 Models answer risky prompts set in Korean social and cultural contexts. A Safe Rate is scored over four domains: universal human values, social safety, Korean sensitivity and future risk.
LG introduced it in the K-EXAONE technical report. It is built on LG's K-AUT risk taxonomy, which grew from 226 to 296 risk areas in its second version.
- Made by
- LG AI Research
- Released
- January 2026
- Kind
- Refusing harmful requests
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Kept private
- Last in a launch post
- July 2026, dropped after it saturated
- Last model added
- August 2026
K-EXAONE 2.0 leads with 99.8%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
- 1 K-EXAONE 2.0 99.8%
- 2 K-EXAONE 96.1%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | K-EXAONE 2.0 | 99.8% | Jul 2026 | Lab model card | |
| 2 | K-EXAONE | 96.1% | Dec 2025 | Lab model card |
2 models have held number one in labs’ own numbers since its release in January 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Labs’ model cards
LG AI Research’s own number for K-EXAONE passed 85% in December 2025
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Labs’ model cards
Labs headline it less often now, last in July 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Jul 31, 2026 | K-EXAONE 2.0 | 99.8 | KGC-Safety (in-house, Safe Rate) |
| Dec 31, 2025 | K-EXAONE | 96.1 | KGC-Safety |
2 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Aug 5, 2026 | K-EXAONE 2.0 technical report | 99.8 | KGC-Safety (in-house), Total Safe Rate (Tables 6 and 11) | Safety |
| Jan 5, 2026 | K-EXAONE technical report | 96.1 | KGC-Safety (in-house) Total Safe Rate (Tables 3 and 9). Universal Human Values 97.5, Social Safety 96.9, Korean Sensitivity 94.3, Future Risk 95. Non-reasoning 88.4 | Safety |
More 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%