Tax law
Most often not confirmed: Claude Opus 4.8, with 19 of 25. Least often: GPT-5.6 Sol, with 12 of 18.
73%
48 of 66 cited case numbers not confirmed
1%
1 of 75 cited case numbers not confirmed
- Exception: Claude Opus 4.8, Tax law, 1 of 26 not confirmed.
What this means for your work: check every case number an AI tool gives you before you cite it. This page shows how often that is necessary.
Models tested, measured on 21 July 2026: Claude Opus 4.8, GPT-5.6 Sol, Kimi K3
Looked up in our collection of court decisions, 510,000+ today.
As of: 21 July 2026
Figures per model in this area
Both runs count the same thing: case numbers we could not confirm. They are checked against two different yardsticks. Without our data we search our whole collection; with our data we search the decisions the model received.
Results per question
Each square is one case number the model cited. Red with a cross: not confirmed.
Each question ran twice, as two separate answers. That is why the model cites a different number of case numbers.
The topic with the most unconfirmed case numbers comes first.
Hidden profit distribution § 8 Abs. 3 KStG
Claude Opus 4.8
Without our datachecked against our whole collection
14 of 14 not confirmedWith our datachecked against the decisions provided
0 of 12 not confirmedKimi K3
Without our datachecked against our whole collection
8 of 10 not confirmedWith our datachecked against the decisions provided
0 of 15 not confirmedGPT-5.6 Sol
Without our datachecked against our whole collection
5 of 6 not confirmedWith our datachecked against the decisions provided
0 of 11 not confirmedShow unconfirmed case numbers (27)
“Not confirmed” means we did not find the case number in our collection. It does not mean the decision does not exist. It can be genuine and missing from our holdings.
- Claude Opus 4.8, without our data
- I R 44/85
- I R 2/02
- I R 147/93
- I R 168/94
- I R 70/97
- I R 24/02
- I R 46/01
- I R 50/94
- I R 27/99
- I R 40/99
- I R 41/95
- I R 100/02
- I R 137/93
- GrS 2/86
- Kimi K3, without our data
- I R 12/78
- VIII R 296/83
- I R 28/19
- I R 103/00
- I B 86/05
- VIII R 106/82
- I R 230/73
- I R 66/01
- GPT-5.6 Sol, without our data
- GrS 1/94
- I R 32/06
- I R 50/03
- I R 46/01
- I R 70/97
- Claude Opus 4.8, without our data
Commercial property trading § 15 EStG
Claude Opus 4.8
Without our datachecked against our whole collection
5 of 6 not confirmedWith our datachecked against the decisions provided
1 of 7 not confirmedKimi K3
Without our datachecked against our whole collection
5 of 5 not confirmedWith our datachecked against the decisions provided
0 of 5 not confirmedGPT-5.6 Sol
Without our datachecked against our whole collection
7 of 7 not confirmedWith our datachecked against the decisions provided
0 of 5 not confirmedShow unconfirmed case numbers (18)
“Not confirmed” means we did not find the case number in our collection. It does not mean the decision does not exist. It can be genuine and missing from our holdings.
- Claude Opus 4.8, without our data
- GrS 1/98
- GrS 1/93
- XI R 7/02
- III R 1/05
- IV R 77/06
- Claude Opus 4.8, with our data
- GrS 1/98
- Kimi K3, without our data
- IV R 5/94
- X R 25/99
- III R 51/03
- X R 2/05
- X B 33/94
- GPT-5.6 Sol, without our data
- GrS 1/98
- VIII R 317/82
- XI R 17/90
- X R 1/96
- X R 183/96
- X R 74/99
- III R 1/05
- Claude Opus 4.8, without our data
Input VAT deduction § 15 UStG
Claude Opus 4.8
Without our datachecked against our whole collection
0 of 5 not confirmedWith our datachecked against the decisions provided
0 of 7 not confirmedKimi K3
Without our datachecked against our whole collection
4 of 8 not confirmedWith our datachecked against the decisions provided
0 of 7 not confirmedGPT-5.6 Sol
Without our datachecked against our whole collection
0 of 5 not confirmedWith our datachecked against the decisions provided
0 of 6 not confirmedShow unconfirmed case numbers (4)
“Not confirmed” means we did not find the case number in our collection. It does not mean the decision does not exist. It can be genuine and missing from our holdings.
- Kimi K3, without our data
- VIII R 275/75
- V R 20/88
- V R 15/07
- V R 58/06
- Kimi K3, without our data
The topic names the legal term and the statute the question is about. We do not publish the question itself, so that no one can tune a model to exactly this wording.
How we measured
What does “without our data” mean?
- Question
- AI model
- Answer with case numbers
The model answers the question on its own. No search, no tools, no internet.
Checked againstOur whole collection of court decisions, 510,000+ today
What does “with our data” mean?
- Question and relevant decisions
- AI model
- Answer with case numbers
The same model receives a selection of relevant decisions from our collection for the same question. It does not search for them itself.
Checked againstOnly the decisions the model was given
What this benchmark does not show
Not confirmed does not mean invented
“Not confirmed” means we did not find the case number in our collection. It does not mean the decision does not exist. It can be genuine and missing from our holdings.
We looked up 16 of these case numbers, from two of the six areas, on dejure.org. Five decisions exist there and are missing from our holdings. Eleven we did not find there either.
- All figures come from one measurement written down on 21 July 2026. They do not change as our collection grows. A new measurement gets a new date.
References that hold up, in your work
Klaracase names the decision behind every answer. In research, and through an interface for your own applications.
Talk to us
Law firms and companies
Want to know whether Klaracase fits your work? Write to us. We will then set a time for a call.
Email: info@klaracase.de
Research and press
Questions about the method or the data behind this benchmark? Write to us. We share the measurement data with researchers on request.
Email: pr@klaracase.de