← series index

# AI for Researchers — Webinar Series Design

**Date:** 2026-07-27
**Status:** Draft for review
**Owner:** Ehsan

## 1. Purpose & Audience

A 7-session webinar series teaching academic researchers (all fields) to use AI effectively and responsibly across the research lifecycle.

- **Audience:** Faculty, postdocs, and PhD students in any discipline. Assumed baseline: casual AI users — they have tried ChatGPT-style tools but use them naively (no systematic prompting, unaware of research-specific tools, policies, or pitfalls).
- **Goal:** Balanced literacy + practice. Every session pairs a conceptual foundation with a hands-on workflow.
- **Positioning:** Fully tool-neutral. Each topic surveys the best current tools (ChatGPT, Claude, Gemini, Elicit, Consensus, SciSpace, NotebookLM, ResearchRabbit, Scite, …). SCiNiTO appears naturally as one option where relevant — no product pitch.
- **Success criteria:** An attendee finishes the series able to (a) explain how LLMs work and where they fail, (b) run a citation-grounded literature workflow, (c) use AI in writing within publisher policy, (d) apply AI to data/code with reproducibility guardrails, and (e) know the specialized AI landscape of their own discipline.

## 2. Format

- 60-minute webinars: ~20 min literacy foundation + ~25 min live workflow demo + ~15 min Q&A.
- Sessions build on each other but each opens with a 2-minute recap so drop-ins can follow.
- 7 sessions (can be compressed to 6 by merging Sessions 2 and 3 if scheduling demands).

## 3. Curriculum

### Session 1 — AI Foundations for Researchers
- Mental model of LLMs: training, tokens, context windows, why hallucination is structural, what "reasoning" models change.
- Capability/failure map for research tasks: what AI is reliably good at, unreliable at, and dangerous for.
- The 2026 tool landscape taxonomy: general chatbots vs. research-specific tools vs. deep-research agents.
- Systematic prompting basics: context, role, constraints, iteration, output evaluation.
- **Demo:** The same research question asked badly vs. well across two tools; live demonstration of a hallucinated citation and how to catch it.

### Session 2 — Literature Discovery & Synthesis
- Semantic vs. keyword search; why AI search finds papers Google Scholar misses (and vice versa).
- Citation-grounded discovery tools: Elicit, Consensus, Semantic Scholar, SCiNiTO, Scite, ResearchRabbit.
- Deep-research agents (ChatGPT / Gemini / Claude deep research): what they do well, where they mislead, how to verify.
- Caution zones: systematic reviews, coverage bias, paywalled literature.
- **Demo:** One real research question run through a discovery tool and a deep-research agent side by side, with citation verification.

### Session 3 — Reading, Notes & Knowledge Management
- AI-assisted reading: paper Q&A / PDF chat; when summaries flatten nuance (critical-literacy segment).
- Grounded notebooks (NotebookLM-style): source-bound Q&A over your own corpus.
- Structured extraction across many papers (methods, effect sizes, populations → comparison tables).
- Integration with reference managers (Zotero and similar).
- **Demo:** Build a grounded notebook from 5 papers, interrogate it, extract a comparison table.

### Session 4 — Writing, Publishing & Integrity
- AI as editor vs. ghostwriter; discipline-appropriate uses, including translation/polishing for non-native English speakers.
- The 2026 policy landscape: ICMJE guidance; the publisher spectrum from prohibition (Science/AAAS) through structured disclosure (Elsevier, Springer Nature, Wiley) to permissive-with-transparency (JAMA); universal rule that AI cannot be an author.
- The disclosure gap as a teaching hook: ~70% of journals have AI policies, yet ~0.1% of papers disclose use.
- Why AI detectors are unreliable; peer-review confidentiality rules; funder policies.
- **Demo:** A policy-compliant editing workflow on a real paragraph + writing a correct disclosure statement.

### Session 5 — Data, Code & Your AI Workflow
- AI for data analysis: code generation, statistical assistance, visualization — with reproducibility guardrails (seed, version, verify).
- No-code analysis paths for non-programmers.
- Agentic coding/analysis tools and where research work is heading.
- Capstone: assembling a personal AI workflow across the whole research lifecycle; how to stay current.
- **Demo:** Dataset → AI-assisted analysis → checked, reproducible result.

### Session 6 — AI for Multidisciplinary Research
- Crossing field boundaries: using AI to translate unfamiliar disciplines' jargon, methods, and core literature.
- Finding adjacent literature and unexpected connections (citation-graph tools, semantic search across fields).
- Cross-disciplinary synthesis: combining evidence from fields with different methods and standards.
- Team science: AI in collaboration — shared knowledge bases, finding collaborators, communicating across expertise levels.
- **Demo:** A researcher from field A uses AI to get oriented in field B: jargon translation → landscape map → annotated reading list → cross-field synthesis memo.

### Session 7 — Discipline Deep-Dives: Specialized AI by Field
A guided tour of specialized (non-chatbot) AI transforming each field, so every attendee leaves with a map of their own discipline:
- **Medicine & clinical research:** diagnostic/imaging AI, clinical-trial matching, PubMed-grounded assistants, regulatory context.
- **Biotechnology & life sciences:** AlphaFold 3 and protein structure, protein/enzyme design, AI drug discovery.
- **Engineering:** simulation surrogates, generative design, CAD and engineering-code copilots.
- **Physical sciences & math:** materials discovery, AI in astronomy/physics pipelines, theorem provers and math assistants.
- **Social sciences & humanities:** large-scale text analysis, AI for qualitative coding, digital humanities, survey/agent-based simulation.
- Closing: how to track your field's AI frontier (key venues, benchmarks, review articles).
- **Demo:** Two short deep-dives chosen by audience poll, run live in field-specific tools.

## 4. Deliverables (per session, built in phase 2)

1. **Slide deck** — ~25–30 slides, HTML-based (exportable to PDF/PPTX, consistent visual identity across the series).
2. **Demo script** — step-by-step walkthrough with exact prompts, expected outcomes, and fallback screenshots in case a live demo fails.
3. **Attendee handout** — 2–3 pages: prompt templates, tool comparison table, checklist, further reading.

## 5. Research & Grounding Methodology

The series must be deeply researched and defensible in front of an academic audience:

- **Deep-research pass per session:** before authoring any session's content, run a dedicated research phase (multiple independent searches: peer-reviewed literature, official policy pages, primary tool documentation, recent benchmark/review articles).
- **Source standard:** every factual claim in slides and handouts must trace to a citable source — peer-reviewed papers, official publisher/funder policy pages, or primary documentation. Vendor blogs and listicles may inform tool discovery but are not citable grounds for claims.
- **`sources.md` per session:** each session folder keeps a dated, annotated reference list; slides carry citation footnotes for statistics and policy claims.
- **Freshness:** all content is date-stamped ("landscape as of <month year>"); a pre-delivery review checks each session's claims shortly before its webinar date, since the field moves quickly.

## 6. Repository Structure

```
ai-research/
├── curriculum-design.md      # this document
├── curriculum.md             # master curriculum (attendee/promo-facing)
├── session-1-foundations/
│   ├── slides.html
│   ├── demo-script.md
│   ├── handout.md
│   └── sources.md
├── session-2-literature/     # …same structure for sessions 2–7
└── assets/                   # shared visual identity, logos, templates
```

## 7. Build Order

1. Master curriculum document (`curriculum.md`) — session descriptions, learning objectives, promotional abstracts.
2. Session content produced one session at a time (deep research → slides → demo script → handout), reviewed by Ehsan before moving to the next.

## 8. Out of Scope

- Recording/hosting logistics, registration, and marketing materials (can be a later phase).
- Hands-on coding workshops (the series stays demo-based; attendees are not assumed to code).
- Translated/localized versions of the content (possible later phase).