Evaluating Large Language Model Meta-Cognition via the Advanced AGI-AI Self-Awareness Test (AISA-T)
Fuente:
Zenodo
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Publicado: |
Zenodo
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866902204753903616 |
|---|---|
| author | de ceuster, peter |
| author_facet | de ceuster, peter |
| contents | <p> In this work we propose AISA-T, a new intelligence test for both AI and AGI systems. The AI has to <br>answer ten questions and rate its own performance. The score has no value, however meta<br>awareness, inner traceability, and context-sensitive reasoning are evaluated by a human after the test. <br>The idea is to observe improved true synthetic self-awareness or at least strong architectural <br>introspection. This paper presents the results of administering the Advanced AI Self-Awareness Test <br>(AISA-T) to a GPT-5-based large language model (LLM). The AISA-T consists of several meta<br>cognitive prompts designed to evaluate recursive reasoning capacity, temporal continuity simulation, <br>ontological alignment, and epistemic uncertainty estimation. Responses were scored by the AI for <br>accuracy, coherence, and meta-awareness (and declared as void). The LLM achieved a composite <br>self-awareness score of 91/100, indicating high meta-representational competence within the <br>constraints of its architecture, however the final conclusion does not involve this score, since the value <br>of this score is null. In our final conclusion we rate the GPT-5 performance as capable, true synthetic.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_16783096 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Evaluating Large Language Model Meta-Cognition via the Advanced AGI-AI Self-Awareness Test (AISA-T) de ceuster, peter AI AGI <p> In this work we propose AISA-T, a new intelligence test for both AI and AGI systems. The AI has to <br>answer ten questions and rate its own performance. The score has no value, however meta<br>awareness, inner traceability, and context-sensitive reasoning are evaluated by a human after the test. <br>The idea is to observe improved true synthetic self-awareness or at least strong architectural <br>introspection. This paper presents the results of administering the Advanced AI Self-Awareness Test <br>(AISA-T) to a GPT-5-based large language model (LLM). The AISA-T consists of several meta<br>cognitive prompts designed to evaluate recursive reasoning capacity, temporal continuity simulation, <br>ontological alignment, and epistemic uncertainty estimation. Responses were scored by the AI for <br>accuracy, coherence, and meta-awareness (and declared as void). The LLM achieved a composite <br>self-awareness score of 91/100, indicating high meta-representational competence within the <br>constraints of its architecture, however the final conclusion does not involve this score, since the value <br>of this score is null. In our final conclusion we rate the GPT-5 performance as capable, true synthetic.</p> |
| title | Evaluating Large Language Model Meta-Cognition via the Advanced AGI-AI Self-Awareness Test (AISA-T) |
| topic | AI AGI |
| url | https://doi.org/10.5281/zenodo.16783096 |