General Use and Scope
-
Can I use artificial intelligence (AI) tools to draft sections of my proposal?
Yes. However, you must fully verify and take responsibility for the accuracy, completeness, and relevance of all content generated by AI. The accountability for the integrity and quality of the final application remains solely with you, the applicant.
-
Is it appropriate to use AI tools to translate parts of my proposal (e.g., from French to English)?
Yes, this is acceptable, particularly to improve clarity. Nonetheless, it remains your responsibility to review and confirm the accuracy of the translated or summarized content.
-
Is it appropriate to use AI tools to summarize parts of my proposal?
Yes. However, be aware of the risks of using online AI tools when you upload your content. See the section below on Confidentiality and Privacy.
Disclosure and Acknowledgement
-
I noticed that the Agencies have removed the mandatory requirement in the Guidance to disclose the use of AI. Why was that?
The Agencies have recognized that a mandatory requirement to disclose AI usage may cause uncertainty, inconsistency, and an unnecessary administrative burden for the research community. Furthermore, monitoring and enforcing comprehensive AI disclosure requirements is resource-intensive and challenging to implement effectively.
-
Does this mean that I do not have to disclose the use of AI at all?
Yes, that is correct—there is no longer a blanket requirement for AI disclosure. However, if an AI tool identifies or surfaces specific works or sources that significantly inform the content of your proposal, you are expected to acknowledge and appropriately reference those original sources. Researchers remain responsible for maintaining transparency, accuracy, and integrity in their submissions.
Verification and Accuracy
-
How can I ensure the accuracy of AI-generated content?
As researchers, your established expertise in verifying sources, methodologies, and applying rigorous scrutiny remains paramount. The approach to verifying AI-generated content should reflect the same thoroughness you apply to all research methodologies. To support this, the following are some best practices in verifying the accuracy of AI-generated content:
- Cross-reference with authoritative sources: Consult peer-reviewed journal articles, textbooks, or reputable databases.
- Consult subject matter experts: Seek validation from colleagues or experts.
- Check citations rigorously: Verify details using bibliographic databases.
- Corroborate concepts and claims: Ensure alignment with recognized findings in your field.
Confidentiality and Data Privacy
-
Are there restrictions on inputting sensitive or proprietary data into AI tools?
Yes. Before using AI tools with sensitive or proprietary data, it is strongly advised that you review the AI platform’s privacy policies, data retention practices, reuse terms, and associated data security controls to ensure that the platform adequately protects your intellectual property and confidential information. Whenever possible, use secure and institutionally approved platforms that explicitly assure your control over data retention and reuse. Ultimately, you remain fully responsible for safeguarding your sensitive and/or proprietary data.
Accountability, Integrity, and Potential Consequences
-
Am I responsible for inaccuracies introduced by AI tools?
Yes. Applicants remain fully responsible for any inaccuracies or unverified content introduced by AI tools.
Policy Updates and Communication
-
How often will AI-related policies be updated, and how will I be informed?
The Agencies will regularly update its policies as AI technologies evolve. Applicants will be informed via official communications and updates to funding guidelines.