Tendril
Loading…
New · guided experience
A curated walkthrough of the library — ordered lessons, a 15-question quiz between each, and 3 next-steps so you stay in flow. Earn XP, badges, and a streak as you go.
Library · 6440 lessons · Research view · NotebookLM
You are viewing research lessons focused on NotebookLM. Use the tool lanes below to jump sideways into related workflows.
Drill down
Start with a real app or workflow. Each lane filters the library to practical lessons, not just broad theory.
603 lessons in research
Lessons handpicked for the NotebookLM shelf.
NotebookLM only answers from PDFs you upload. The teen study trick that gives you AI without the hallucinations.
Upload a PDF, a set of docs, or a research paper. NotebookLM produces a two-host podcast conversation about the material.
NotebookLM turns your documents into an AI tutor that only answers from your sources. Look at why its audio overviews went viral and where it still falls short.
NotebookLM is Google's AI that ONLY answers from documents YOU upload — perfect for studying.
Fresh NotebookLM lessons added to the library.
NotebookLM only answers from PDFs you upload. The teen study trick that gives you AI without the hallucinations.
NotebookLM is Google's AI that ONLY answers from documents YOU upload — perfect for studying.
NotebookLM turns a pile of PDFs into a searchable, askable brain. Here is how to build a research notebook that keeps paying dividends.
NotebookLM turns your documents into an AI tutor that only answers from your sources. Look at why its audio overviews went viral and where it still falls short.
Browse everything
Subject tracks
Tap a tile to filter the library — or pick “Surprise me” below for a randomized starter set.
Detection arms races don't produce honest students. AI literacy education — helping students understand what counts as their own thinking and why — is the only approach that survives the next generation of AI tools.
AI can give students fast feedback on essays — comma usage, structure, argument strength. The art is using it to deepen teaching, not deskill teachers.
AI converts a chronological account into a structured incident narrative focused on system factors.
AI is a useful reflection partner for burnout, not a substitute for a therapist or your peer-support program.
AI compares the true 4-year cost of 5 colleges so you pick on math, not vibes or rankings.
AI can summarize a company's basics so you actually understand what you're buying — it can't predict the next moonshot.
AI did not start in 2022. It has decades of wrong turns and breakthroughs. Knowing the history helps you spot hype from real progress.
P-value is one of the most abused numbers in research. Here is what it actually says — and what it does not. 'Model B is no better than model A.' 'The new prompt does not change user satisfaction.' A low p-value means the boring story would rarely produce data that looks like what you saw.
Not all hallucinations are alike — citation lies, fact lies, and confident-tone lies each need a different defense.
Apps like Woebot use AI to help with everyday stress and feelings. Useful for some stuff. Not a replacement for a real therapist or trusted adult.
AI compares acne treatments based on actual research, not TikTok hype.
Not all period apps treat your data the same. AI can compare them so you don't have to read 9 privacy policies.
AI can make scary-real deepfakes — using them on classmates can get you charged with felonies, not just suspended.
OpenAI, Voyage, Cohere, and open-source models all do embeddings — best one depends on your use case.
Wikipedia gets a bad rap in school, but it's still one of the best places to start a research project. The trick is knowing how — not whether — to use it. But the rule is more nuanced than "never use it." Smart researchers — including AI researchers — start at Wikipedia and use it as a launchpad to better sources.
When AI mentions a study, book, or article, your job is to verify the source actually exists — not just trust AI's summary of it.
Taking notes by copy-pasting AI summaries doesn't help you learn. Note-taking is most powerful when you put ideas into your own words — which forces real understanding.
Paraphrasing is putting an idea in your own words after you understood it. Word-swapping is just sneaky copying. Schools detect both — but only one is real research. "AI is helpful" becomes "Artificial intelligence is useful." That's not paraphrasing — that's sneaky copying.
AI as a research coach asks you good questions, points out weak spots, and helps you think clearer. AI as a ghostwriter does your work for you. Same tool, very different uses.
Vague prompts get vague answers. The skill of research with AI is in the question, not the tool.
Not every source on the internet is reliable. AI helps you evaluate credibility before citing.
Dump your class notes into AI and get back a clean, organized study guide in minutes.
AI invents stats with confidence — here's where to find numbers you can actually cite.
Use AI to find which actual scientists and researchers post on social — then follow them, not influencers.
NotebookLM only answers from PDFs you upload. The teen study trick that gives you AI without the hallucinations.
AI search personalizes — meaning your feed and answers may not match your friend's, and that shapes what you believe.
Switching from 'search and copy' to 'investigate and synthesize.'
Upload a PDF, a set of docs, or a research paper. NotebookLM produces a two-host podcast conversation about the material.
NotebookLM turns your documents into an AI tutor that only answers from your sources. Look at why its audio overviews went viral and where it still falls short.
NotebookLM is Google's AI that ONLY answers from documents YOU upload — perfect for studying.
Leaderboards are compelling. They are also deeply misleading. Here is a checklist for real skepticism. In reality, leaderboards hide a stack of choices that can swing the ordering: prompt wording, sampling settings, number of attempts, which subset of the benchmark is reported.
NotebookLM turns a pile of PDFs into a searchable, askable brain. Here is how to build a research notebook that keeps paying dividends.
Compare reinforcement learning from human feedback and direct preference optimization at the level of intuition, not equations.
Ollama, LM Studio, and most local-model apps are wrappers around llama.cpp. Knowing what it actually does — and how to drop down to it — pays off when defaults are not enough.
OLMo is valuable because it centers openness: students can discuss not only weights, but data, training recipes, and research reproducibility.
Compare model families on full-task cost including retries and context.