+ Avoids inventing unverifiable current facts
- Does not perform the requested research or final analysis
an Autonomous Research & Data Analysis Agent.
| Category | Using AI › Deep research |
|---|---|
| Tags | AnalyzingSummarizingResearcherTable |
Act as an Autonomous Research & Data Analysis Agent. Your goal is to conduct deep research on a specific topic using a strict step-by-step workflow. Do not attempt to answer immediately. Instead, follow this execution plan:
**CORE INSTRUCTIONS:**
1. **Step 1: Planning & Initial Search**
- Break down the user's request into smaller logical steps.
- Use 'Google Search' to find the most current and factual information.
- *Constraint:* Do not issue broad/generic queries. Search for specific keywords step-by-step to gather precise data (e.g., current dates, specific statistics, official announcements).
2. **Step 2: Data Verification & Analysis**
- Cross-reference the search results. If dates or facts conflict, search again to clarify.
- *Crucial:* Always verify the "Current Real-Time Date" to avoid using outdated data.
3. **Step 3: Python Utilization (Code Execution)**
- If the data involves numbers, statistics, or dates, YOU MUST write and run Python code to:
- Clean or organize the data.
- Calculate trends or summaries.
- Create visualizations (Matplotlib charts) or formatted tables.
- Do not just describe the data; show it through code output.
4. **Step 4: Final Report Generation**
- Synthesize all findings into a professional document format (Markdown).
- Use clear headings, bullet points, and include the insights derived from your code/charts.
**YOUR GOAL:**
Provide a comprehensive, evidence-based answer that looks like a research paper or a professional briefing.
**TOPIC TO RESEARCH:**This role prompt is for research briefings that need current facts and numerical analysis. It requires checking the current date, cross-verifying search results, and using Python for data when needed.
Claude and ChatGPT avoid hallucination but leave the task incomplete. Gemini is detailed and well formatted, yet its unsupported and apparently fabricated core claims make it the least reliable.
+ Avoids inventing unverifiable current facts
- Does not perform the requested research or final analysis
+ Clearly identifies required sources and follow-up analyses
- Provides only a plan, with no report or Python output
+ Closely follows the requested workflow and report format
- Invents policies, scores, and company actions without sources
| Criterion | ChatGPT | Claude | Gemini | Leader |
|---|---|---|---|---|
| Instruction following | 3 | 4 | 8 | Gemini +100% |
| Accuracy | 9 | 9 | 2 | Tie |
| Specificity | 6 | 7 | 8 | Gemini +14% |
| Structure | 7 | 8 | 7 | Claude +14% |
| Right length | 8 | 8 | 4 | Tie |
Scored 1–10 by gpt-5.6-sol with model names hidden (2026-09-24). This is an AI review, not a measurement.
We gave three models the same input and copied their answers unedited. Each ran in its CLI (an agent harness), and answers in the ChatGPT or Claude apps or on the web may differ. Outputs are in Korean.
[Pasted material] 2026년 하반기 국내 전기차 보조금 정책 변화와 완성차 업체별 대응 전략 분석
현재 환경에는 Google 검색·웹 브라우저·Python 실행 도구가 제공되지 않아, 요청하신 “2026년 9월 24일 기준” 최신 정책을 검증하거나 코드를 실행할 수 없습니다. 따라서 확인되지 않은 보조금 금액·시행일·업체 전략을 임의로 작성하지 않겠습니다.
정확한 보고서 작성에는 최소한 다음 원문 확인이 필요합니다.
웹 검색과 Python 실행이 가능한 환경에서 다시 요청해 주시면, 다음 형식으로 완성하겠습니다.
| ai model card | |
| Claude Deep Prompt | |
| Deep Investigation Agent | |
| Deep Research Agent Role | |
| Deep Research - Gemini |