Call/WhatsAppText +1 (302) 613-4617

Computer Science

How to Approach the Perplexity AI Discussion Post

WHAT TO TRY  ·  HOW TO REFLECT  ·  WHAT TO WRITE  ·  PERPLEXITY AI  ·  500 WORDS

Perplexity AI Discussion Post

Your professor wants 500 words on how you used Perplexity AI — not a product review, not a formal essay. This guide breaks down what the assignment is actually testing, what to try in the tool, and how to write a reflection that sounds like a real person rather than a Wikipedia entry.

8–10 min read AI / ML Tools AI-Powered Search Discussion Post

Need help writing your AI tools reflection post? Our academic writing team can guide you through it.

Get Help →
Custom University Papers — Academic Writing Team
Guidance for AI/ML tools assignments. References Perplexity AI and its published documentation on search and citation methodology.

Most students approach this assignment backwards. They spend an hour reading about Perplexity AI and five minutes actually using it. Then they write a description of what the tool does rather than a reflection on what they experienced. The assignment explicitly says it isn’t a writing assignment — it wants your process and your honest take. So the more you actually use the tool before you write, the better your post will be.

What Perplexity Actually Does Choosing Your Topic Prompts Worth Trying What Worked vs What Didn’t How to Write the 500 Words Mistakes That Lose Points Perplexity vs Other AI Tools

What Perplexity AI Actually Is — Before You Try to Use It

Perplexity launched in 2022 and describes itself as an “answer engine” rather than a search engine. The distinction matters for this assignment. Regular Google returns a list of links. Perplexity returns a direct answer — written in natural language — with citations embedded inline so you can see exactly which source each claim comes from.

The Core Mechanism

LLM + Live Web Search, Not Just a Chatbot

Perplexity combines a large language model (it uses multiple, including GPT-4 and Claude depending on settings) with real-time web search. When you ask a question, it searches the web, pulls relevant content from live sources, and synthesizes that into a structured answer — with numbered citations you can click to verify. This is different from ChatGPT, which generates answers from training data without live search by default.

Why this matters for your assignment: The citation mechanic is what makes Perplexity distinctive. A lot of your reflection can focus on whether the citations actually supported what Perplexity claimed, whether they were credible sources, and whether you trusted the answers more or less because of them.
2022 Year Perplexity AI Launched
Live Web Search Built Into Every Answer
Cited Every Claim Sourced With Inline Citations
The Free Tier Is Enough for This Assignment

You don’t need Perplexity Pro. The free version gives you full access to the core search-and-answer functionality, follow-up questions, and source viewing. Pro adds more advanced model options (like GPT-4o) and higher usage limits, but the base experience — which is what the assignment is asking you to evaluate — is entirely free. Go to perplexity.ai and start there.

How to Pick Your Topic

The assignment says “choose a topic that you are interested in.” That’s not a throwaway line. Pick something you actually know something about — your major, a course you’re taking, a hobby, a news story you’ve been following. Why? Because when you already have some background knowledge, you’ll notice immediately when Perplexity gets something wrong or oversimplifies. That friction is gold for your reflection.

Topics That Generate Interesting Reflections

  • Your major’s current debates — Are there contested findings in your field? Perplexity’s handling of nuanced academic disputes is worth testing.
  • Recent news events — Live search means Perplexity can pull in things that happened last week. How current are its answers?
  • Technical how-to questions — Ask it to walk through a process you know well. Does it get the steps right?
  • Comparative analysis — “Compare X and Y” queries show how the tool handles trade-offs and ambiguity.
  • Literature or film analysis — Does it engage with interpretation or just summarize plot?

Topics That Make the Post Harder to Write

  • Hyper-niche topics — If Perplexity can’t find good sources, you’ll get thin answers and there’s less to reflect on.
  • Pure opinion questions — “What’s the best movie ever?” won’t generate much. The tool isn’t built for opinion.
  • Topics you know nothing about — You can’t evaluate accuracy if you have no baseline.
  • Anything that requires proprietary data — Internal company reports, paywalled journals — Perplexity can’t pull those and the gaps won’t be interesting to write about.

Specific Things to Try in the Tool

Don’t just type one query and call it a session. The assignment asks you to use Perplexity “meaningfully” — which means exploring its features, not just testing one prompt. Here’s a structured way to get enough material to write 500 good words.

Session Structure

Spend 20–30 Minutes Actually Using It

Start with a broad question on your topic. Read the answer carefully — not just the main text, but the citations. Click at least two or three of the source links and check whether the citation actually says what Perplexity claims it says. Then ask a follow-up question using the thread feature. Change the “Focus” mode (Academic, YouTube, Reddit, etc.) and see how the answer changes. Finally, ask a question where you already know the answer and check whether Perplexity gets it right.

The thread feature matters. Perplexity keeps conversational context within a search thread — you can ask “give me more detail on that second point” and it understands what you mean. This is different from typing into a new search each time. Try the conversation mode, not just single queries.

Prompt Types That Generate Good Reflection Material

These aren’t just good queries — they’re prompts that will give you something concrete to write about in your reflection.

Prompt Type 1 — Broad Factual Question

What are the current leading theories on [topic in your field]?

What to note: How current are the sources? Does it acknowledge disagreement between researchers, or flatten everything into a single “the consensus is…”?
Prompt Type 2 — Contested Question

What are the strongest arguments for and against [position you care about]?

What to note: Does Perplexity present both sides fairly? Does it hedge too much, or take a side? Are the cited sources actually credible on this topic?
Prompt Type 3 — Verification Test

Ask it something you already know the answer to from class or personal experience.

What to note: Did it get it right? If it was wrong or oversimplified, was the error in the LLM’s synthesis or in the sources it pulled? This is one of the most interesting things to reflect on.
Prompt Type 4 — Academic Focus Mode

Switch to “Academic” focus and search for recent research on [your topic].

What to note: Does Academic mode pull peer-reviewed papers? How does the quality of sources compare to regular mode? Would you use this for actual research?
Prompt Type 5 — Data or Statistics Request

What does recent data show about [measurable aspect of your topic]?

What to note: Does it cite primary data sources (government databases, peer-reviewed studies) or secondary summaries? Can you verify the numbers?

How to Structure the 500 Words

The assignment says it’s not a writing assignment — but that doesn’t mean structure doesn’t matter. A disorganized stream of thoughts won’t score as well as a reflection that has a clear arc. Here’s a simple structure that works without feeling like a formal essay.

What to Cover in 500 Words

Open with your topic and why you picked it — One sentence. “I decided to use Perplexity to research X because I’m currently taking a course on Y and wanted to see how it handles contested questions in that area.” That’s it. Don’t over-explain.
Explain what Perplexity is in 2–3 sentences — What makes it different from a regular search engine or a standard chatbot. Focus on the citation mechanic and the live web search. Don’t write a product description — just enough context so your reflection makes sense.
Name your actual prompts — Quote at least two or three of the queries you ran. Specificity here is what separates a real reflection from a generic one. “I asked it X, and it responded by…” is always better than “I explored the tool and found it useful.”
Talk about what worked — What did Perplexity do well? Did the citation feature help? Was the synthesis accurate? Was the answer more useful than a Google search would have been? Give a specific example.
Talk about what didn’t work — This is where most students go weak. They don’t want to criticize the tool. But your professor wants honest evaluation. Did Perplexity miss nuance? Cite a poor source? Get something factually wrong? Surface that. It shows you thought critically.
End with your takeaway — Would you use it again? For what kinds of tasks? How does it compare to other tools you’ve used? One honest paragraph. No need to wrap up with a grand conclusion.

Perplexity vs. Other AI Tools — How to Frame the Comparison

The assignment asks you to engage with Perplexity specifically, but your reflection will be stronger if you can place it in context. Here’s a comparison table that might inform how you talk about it.

Tool How It Answers Citations Best For Main Limitation
Perplexity AI LLM synthesis of live web search results Inline citations with source links Research questions that need current, cited sources Synthesis can still hallucinate; citations sometimes don’t fully support the claim
ChatGPT (default) LLM from training data; no live search by default None unless you use Browse mode Writing, brainstorming, code generation Knowledge cutoff; can’t pull real-time information without search plugin
Google Search Ranked list of links; AI Overview in some queries Links to full articles Finding specific sources, navigating to primary documents Requires more work to synthesize across sources; ads affect ranking
Google Scholar Database of academic papers, no synthesis Full paper citations in standard formats Finding peer-reviewed research No natural language interface; requires you to read and synthesize yourself
Don’t Confuse “Cited” With “Correct”

Perplexity’s biggest selling point is also its biggest trap for students. Because every claim has a citation, it feels authoritative. But the LLM can still misrepresent what a source says — it might cite an article that discusses X as evidence for Y, when the article actually argues something more nuanced. Click the citations. Don’t assume that because a source is linked, the claim it’s attached to is accurate. This is worth a paragraph in your reflection — and it’s a genuinely important point about how AI tools can create false confidence through the appearance of sourcing.

Mistakes That Cost Points

Writing About What Perplexity Is Instead of What You Did With It

The assignment says “tell us what prompts you tried, how you engaged with the tool.” A product description is not a reflection. If your post reads like a Wikipedia summary of Perplexity’s features, you’ve missed the point.

Lead With Your Experience, Add Context as Needed

Open with what you did. “I used Perplexity to research X. My first query was Y. Here’s what happened…” Background on the tool should support your reflection, not replace it.

Staying Too Positive — All Praise, No Critique

Every student says the tool was “very useful” and “easy to use.” That’s not evaluation, that’s politeness. Your professor knows the tool has limitations. Show you noticed them.

Call Out One Specific Failure

Find one moment where Perplexity fell short — a misleading citation, an oversimplified answer, a query it couldn’t handle well — and describe it specifically. That’s the most credible thing you can put in a 500-word reflection.

Using Vague Language: “I explored the tool and found it useful for research”

This tells the reader nothing. What did you search? What did you find? What surprised you? Vague language signals that you didn’t engage deeply with the tool.

Quote Your Actual Prompts and Describe Actual Outputs

“When I asked ‘what are the main critiques of behavioral economics,’ Perplexity pulled 6 sources, three of which were from peer-reviewed journals and two from news outlets. The answer was accurate but didn’t engage with the methodological debates I’d read about in class.” That’s a real reflection.

Forgetting to Attach Outputs or Links

The assignment says “submit links / attach documents to any relevant projects, reports, or outputs.” Many students forget this. Perplexity lets you share search threads via a share link. Copy and paste a few of them.

Share the Actual Threads

After each query session in Perplexity, click the share button and copy the URL. Paste those links into your submission. This takes 30 seconds and satisfies the “attach documents” requirement directly.

Frequently Asked Questions

What’s the difference between Perplexity AI and ChatGPT?
The core difference is live web search with inline citations versus generation from training data. ChatGPT generates answers from what it learned during training (with a knowledge cutoff, and no live sources unless you use its Browse feature). Perplexity runs a live web search every time you ask a question and shows you exactly which sources each part of its answer comes from. That makes Perplexity more reliable for current events and verifiable facts, but it also means you can check its work — and should. Neither tool is always right. Perplexity’s answers are more auditable because of the citations, but the synthesis step can still introduce errors.
What does “Focus” mode do in Perplexity?
Focus modes let you restrict where Perplexity searches. “All” searches the open web. “Academic” restricts to academic papers and journals, pulling from sources like PubMed, arXiv, and Semantic Scholar. “YouTube” finds relevant video content. “Reddit” pulls from community discussions. For your assignment, try switching between at least two modes on the same query and compare the answers — the difference between Academic mode and general web mode on a research question is usually stark and worth writing about.
How do I share my Perplexity session for the assignment?
After running a query or a thread of queries, look for the share icon at the top of the thread — it looks like an arrow pointing out of a box. Click it, and Perplexity generates a shareable URL that preserves the full conversation including sources. Anyone with that link can view your thread without needing a Perplexity account. Paste those links directly into your assignment submission. If your submission is a document, include them as footnotes or in an appendix section at the end.
Do I need a Perplexity Pro account to complete the assignment?
No. The free tier handles everything this assignment requires. You can run queries, follow up within a thread, switch between focus modes, view and click citations, and share threads — all without paying. Pro adds access to more powerful underlying models and higher usage limits, but the fundamental experience — which is what you’re evaluating — is identical. Sign up with a Google or email account and start immediately.
What if Perplexity gets something factually wrong in my session?
Write about it. That’s the most valuable material you can include in your reflection. Describe the query, explain what Perplexity said, explain why it was wrong (and how you know — either from your course material, a primary source, or by clicking the citation and seeing that the source didn’t actually say what Perplexity claimed). AI tools getting things wrong is not a problem for your assignment — it’s content. The professor isn’t grading you on whether the tool worked perfectly; they’re grading you on whether you engaged critically with it.
Can I use Perplexity to research other AI tools for the assignment?
Yes, and it’s actually a good move. Ask Perplexity to explain how it works, what LLMs it uses, how its citation system functions, or how it compares to competitors. This serves two purposes: you’re using the tool to research itself (which is a form of “meaningful use”), and you get material for the comparison section of your reflection. Just make sure the bulk of your usage is on a substantive topic — using Perplexity exclusively to research Perplexity might feel circular to a grader.

One Last Thing Before You Open the Tool

The 500-word limit is tight. Most students write 600 and then cut — which is the right approach. Write freely first. Get your actual experience down. Then trim the generic parts and keep the specific ones.

The strongest posts your professor will read this semester will have three things in common: specific prompts quoted verbatim, at least one honest criticism, and a clear sense of what the student would actually use the tool for going forward. Everything else — background on Perplexity, feature descriptions, comparisons — is supporting material.

Don’t try to sound like you know more about AI than you do. The assignment is asking how it felt to use the tool, what surprised you, and what you’d do differently. That’s a much lower bar than it seems — and it’s one that rewards honesty over performance.

Need Help With Your AI Tools Discussion Post?

Our academic writing team can help you structure your reflection, polish your analysis, or guide you through the entire assignment from first prompt to final submission.

Discussion Post Help Get Started

AI Tools Assignments and Discussion Posts

Academic guidance for AI/ML tool explorations, reflection posts, and technology assignments across disciplines.

Discussion Post Help
Article Reviewed by

Simon

Experienced content lead, SEO specialist, and educator with a strong background in social sciences and economics.

Bio Profile

To top