Manage your consent settings, view session logs, and complete surveys.
Use the same email as your Claude account.
Principal Investigator: Joseph Austerweil, Senior Researcher, Henkaku Center, Chiba Tech
Contact: joseph.austerweil@chibatech.ac.jp
Research Period: April 7, 2026 – March 31, 2029
This study investigates how students develop AI collaboration skills over the course of a semester of structured instruction. As AI tools become a routine part of education and professional practice, understanding how students learn to work effectively with these tools is increasingly important.
The study examines what kinds of AI students engage in naturally as part of their coursework, what patterns of interaction are associated with better learning outcomes, and what potential side effects may arise from sustained AI use in the classroom.
If you consent, your anonymized session logs, survey responses, and structured reflections may be used for research analysis and academic publication, in addition to the normal instructional use that applies to all enrolled students.
Participation involves completing research surveys at the beginning, middle, and end of the semester. These include measures of AI tool familiarity, confidence in working with AI systems, cognitive style, and metacognitive self-regulation. The total additional time commitment is approximately 45–60 minutes across the semester (roughly 25–30 minutes at Session 1 and 15–30 minutes spread across later sessions). In addition, your existing coursework data (session logs and course-embedded reflections) may be used for research purposes.
What session logs contain: Prompts you submit to the AI, responses generated by the AI, timestamps, token counts, and tool invocations (e.g., file edits, code execution).
What session logs do not contain: Audio, video, or screen recordings; keystroke or input data outside of AI interactions; browsing history or activity in other applications; physiological or biometric data.
Surveys and reflections: The research surveys cover topics such as AI tool familiarity, task delegation judgment, self-efficacy in specifying intent to AI, metacognitive awareness of AI collaboration, metacognitive self-regulation, cognitive reflection, need for cognition, and intellectual humility. Some course-embedded reflections (e.g., prompt snapshots, mid-semester check-ins, peer ratings) are completed by all students as part of normal coursework; for participants, these may also be used for research purposes.
All students, regardless of participation, receive a Personal Insights report showing their own interaction patterns (session frequency, duration, prompt characteristics).
Students who consent to participate additionally receive a Cohort Insights report: anonymized aggregate patterns from peers that provide context for interpreting your own data, and connections between your survey responses and AI usage patterns. These features are only possible through the research analysis your consent enables.
Your instructor will not know whether you have chosen to participate. Your decision has no effect on your grade or your experience in this course.
Students who do not consent receive the same instruction, AI tool access, feedback, and grades as participants. You may change your mind and opt in at any later point during the semester. You will never be identified as a non-participant to instructors, peers, or anyone outside the research team.
The principal investigator for this study is also a course instructor. To protect you from any perceived pressure, the following structural safeguards are in place:
Upon logging in to the CLI tool, the system generates a random anonymous identifier for you. This identifier is stored in a signed research token on your machine. The link between your email and your anonymous identifier is stored encrypted on the server — the server does not store any plaintext mapping between your anonymous identifier and your name, email, or university ID.
You use your research token to sync session data and to access your insight reports. Research data and your identity are stored on separate systems. The research system contains no names, student IDs, or other personally identifying information. All analysis, publication, and dataset sharing uses anonymous identifiers exclusively.
Data retention: Research data is retained for five years from the end of the research project, consistent with Chiba Tech institutional policy, and then deleted. The encryption key for the sealed identity mapping is permanently destroyed one month after the end of classes, at which point all retained data becomes irreversibly anonymized.
Small cohort protections (APS-I): For the APS-I cohort (6–15 students), additional protections are used to prevent re-identification: suppression of cells with fewer than 5 observations, composite descriptions combining similar cases, and aggregation across multiple cohorts where possible.
Incidental sensitive content: Session logs capture the full content of your interactions with AI tools. If you discuss personal matters with the AI during a coursework session, that content will appear in the logs. You are advised to be mindful of what you share. If a researcher encounters clearly sensitive personal content (e.g., health information, relationship difficulties, financial hardship), it will be permanently deleted.
If, during the course of research analysis, a researcher encounters information in session logs suggesting a serious safety concern (e.g., risk of self-harm or abuse), the research team will follow Chiba Tech's established student welfare reporting procedures.
You may withdraw your consent at any time during the course and up to one month after the end of classes, with no disadvantage of any kind.
When you withdraw, your research data is immediately hidden from all queries. Data still flagged as withdrawn one month after the end of classes is permanently deleted along with the identity mapping. Your grades, AI tool access, and instruction are not affected.
Withdrawal deadline: One month after the end of classes, the sealed identity mapping is permanently destroyed. After this date, withdrawal is not possible because no person or system can determine which anonymous record belongs to which individual. You will be informed of this deadline at the time of consent and reminded again at the end of the course.
How to withdraw: Toggle Research-use off on this page, or contact the data custodians (Grisha Szep or Fiza Razik) directly.
Benefits: Access to the Cohort Insights report described in Section 3. The study results may also contribute to improved AI education practices for future students at Chiba Tech and beyond.
Risks: The primary risk concerns data privacy. Although your data is anonymized using anonymous identifiers, no system can entirely eliminate the possibility of re-identification — particularly in the smaller APS-I course, where distinctive project topics could make individuals recognizable even without names. This risk is mitigated through encrypted anonymization, permanent destruction of the identity mapping key, and the additional small-cohort protections described in Section 6.
Compensation: There is no monetary compensation, gift cards, course credit, or grade adjustment for participating.
Results will be published in academic journals and presented at domestic and international conferences. De-identified data will be made available through open-science repositories (e.g., Open Science Framework, Zenodo) without identifying specific individuals. No individual student's data will be identifiable in any publication or dataset.
Intellectual property rights arising from this research will be held by Chiba Tech and the researchers. Participants do not acquire intellectual property rights from this research.
This research has been reviewed by the Chiba Tech Ethics Review Committee for Research Involving Human Subjects and approved by the university president. The research is supported by the university budget and grants from the private sector. There are no conflicts of interest with any particular company.
Principal Investigator
Joseph Austerweil, Chiba Tech, School of Design & Science, Henkaku Center
2-17-1 Tsudanuma, Narashino City, Chiba Prefecture
Email: joseph.austerweil@chibatech.ac.jp
Co-Investigators / Data Custodians
Grisha Szep (Researcher, Henkaku Center) · Fiza Razik (Visiting Researcher, Henkaku Center)
Ethics Review Committee
Chiba Tech Ethics Review Committee for Research Involving Human Subjects
Your synced Claude Code sessions, grouped by project.