3 STEP PROCESS OF THESIS WRITING WITH AI TOOLS | AI TOOLS FOR RESEARCHER

Academic writing with AI tools works best as a workflow, not a one-click shortcut to a submission-ready thesis. The core message is that researchers should adopt a multi-step process and combine multiple tools—some free, some paid—because no single AI system can reliably generate a complete, high-quality research document without human judgment, editing, and verification.

The first step centers on Google AI Studio, positioned as an all-purpose writing and thinking assistant for research. It can help generate ideas and transform information across formats. Examples include turning research papers stored in Google Drive into a literature review, extracting information from uploaded files, converting recorded audio into a document, and using images captured by a camera to produce text. It can also convert a YouTube link into a written blog or document, effectively translating the content of a video into text.

Beyond drafting text, Google AI Studio supports planning and structure. Users can start from a rough research topic and refine it by providing details about the research problem, research questions, and intended methodology. From there, it can help identify a research gap, develop aims and objectives, and generate an outline for the thesis or research proposal. The outline is described as more detailed than what some paid alternatives produce, including chapter-by-chapter structure and guidance on what information to include or exclude.

The outline output can be copied and pasted into Google Docs for editing, allowing the researcher to adjust wording, scope, and section organization. The emphasis remains on using AI to reduce the “writing block” and accelerate planning, while keeping control of the final academic narrative.

The second step focuses on literature gathering and citation support using site.ai. This tool is presented as a research assistant for searching sources, brainstorming, and drafting topic-specific background. In practice, it can retrieve literature for specific claims (the transcript gives an example about historical domains of solvents in the chemical industry) and provide citations and reference lists alongside the generated content. Citations are clickable, letting users trace statements back to the source papers, which supports fact-checking and improves the reliability of what gets written.

site.ai also offers ways to shape the retrieved material: responses can be expanded, results can be displayed as tables, and users can switch between search modes or prioritize recent references. The workflow ends with writing and compilation, after which AI-generated text may be “humanized.” The transcript recommends using an AI humanizer tool—specifically mentioning “Hicks bypass”—to rephrase content and avoid triggering AI-detection or “AI plagiarism” checks, with the claim that rewording helps if any issues appear.

Overall, the process is framed as: plan and outline with Google AI Studio, research and cite with site.ai, then revise/humanize before submission. The tools are treated as accelerators for gathering information, building structure, and improving drafts—while the researcher remains responsible for accuracy, originality, and final edits.