According to Tom’s Guide, a head-to-head test of Perplexity AI and Google’s NotebookLM on five complex research prompts resulted in a narrow 3-2 victory for Perplexity. The test involved detailed queries on topics ranging from houseplant care in low-light apartments to evaluating public transit apps in Bristol, UK. NotebookLM, described as an “overly keen teacher,” often provided overwhelming detail from up to 23 sources, including AI-generated videos and podcasts. Perplexity’s Deep Research feature, acting like a “super-efficient research assistant,” tended to deliver more concise, structured answers with clear citations. The overall winner was Perplexity, praised for its focus on actionable details over NotebookLM’s sometimes distracting “flare.” However, NotebookLM won specific rounds where its deep multimedia approach was more effective.
The Style vs. Substance Tug-of-War
Here’s the thing about this test: it reveals a fundamental split in how AI tools are approaching the “research assistant” role. NotebookLM seems built for the user who wants to fall down a rabbit hole. It throws PDFs, diagrams, videos, and a mountain of sources at you. That’s great if you’re starting a long-term project and need to immerse yourself. But for getting a clear, direct answer? It can be too much. I mean, who wants a podcast when they just asked for a list of 10-minute snacks?
Perplexity, on the other hand, is optimized for the user who wants the report, not the workshop. It cuts to the chase with summaries, tables, and sourced evidence right up front. It’s prioritizing answer delivery over exploratory learning. The test shows that NotebookLM’s strength—context and multimedia—can actually become its weakness, causing it to miss the prompt’s core request entirely, like with the snacks or the Bristol transit app. That’s a pretty big flaw.
When More Is Actually Less
The most telling failures for NotebookLM were on the specific, actionable prompts. Asking for the best transit app in Bristol and getting a general report on UK and Singapore data systems? That’s a miss. Requesting simple snack recipes and receiving a treatise on “Satiety and Strategic Snacking”? Another miss. It’s like asking for the time and getting a lecture on the history of clockmaking.
This highlights a critical challenge for AI: understanding user intent beyond the literal keywords. Perplexity seemed to nail this more consistently. It provided a ranked answer for Bristol, complete with app names and reasons. It listed actual snacks with ingredients. Basically, it did the job. In a world where everyone’s pressed for time, that reliability is huge. NotebookLM feels like it’s showing off its capabilities, while Perplexity is focused on solving your problem.
The Future of AI Research Tools
So where does this leave us? I think we’re seeing the early formation of two distinct product categories. One is the deep-dive project companion (NotebookLM), and the other is the precision research engine (Perplexity). The winner in any situation depends entirely on what you’re trying to do.
But for most people, most of the time, getting a correct, concise answer is the priority. That’s why Perplexity’s narrow win here feels significant. It suggests that for now, utility beats spectacle. The trajectory, though, will be for these tools to learn from each other. Can Perplexity add more engaging, digestible formats without losing its focus? Can NotebookLM rein in its enthusiasm and better match its output to the user’s immediate need? The company that successfully blends depth with directness will probably take the crown.
