PaperBrain
PaperBrain is a web platform for exploring scientific literature. Researchers search arXiv, upload their own PDFs, chat with a paper, and turn it into a multi-voice audio podcast. Access starts with a free account.
What is PaperBrain?
PaperBrain is a web platform for exploring scientific literature, online since 2024. Its homepage promise is blunt: Scientific Literature, Simplified for You. Four building blocks sit behind it, all announced on the same page: smart searches, audio podcasts made from papers, easy uploads, and interactive chats with a paper. The corpus is mainly arXiv. The site's own FAQ says PaperBrain 'sources its papers from a variety of sources, mainly through arXiv', and that users can add their own research papers by uploading them. That orientation shows in the audiences it fits: physics, computer science and AI, mathematics, statistics, quantitative biology. The application sits entirely behind a login handled by the Kinde authentication provider. Only the homepage, the blog, the legal pages and shared podcast pages are public. The sitemap exposes the working routes: /search, /upload, /ask, /podcast, /dashboard, /billing and /pricing. The podcast feature is what sets the tool apart. A paper is replayed as a spoken dialogue between named voices, among them Emma Anderson, Ethan Sullivan, Olivia Morgan, Alexander Pierce, Noah Bennet and Stuart Montgomery. The same paper can be rendered at three registers, visible in the podcast URLs as funny, moderate and expert. A public paper page shows the title, the authors, the full abstract, a Transcript section and Play, Share and View PDF buttons. One published example runs on arXiv paper 2408.08535, 'CommunityKG-RAG', by Rong-Ching Chang and Jiawei Zhang. The audio is served from an S3 bucket named paperbrain-podcasts in the ap-south-1 region. The site claims an audience of 100k+ researchers and shows a logo strip that includes the Indian Institute of Technology Delhi. Its declared keywords name the underlying stack: chatgpt, openai, gpt, llm, langchain. PaperBrain is the work of one person, founder Md Anzar Ahmad, presented on a 'Meet the Team' page. Use is free with an account, and a paid plan is described as a way to support development, though no price is published. The limits are equally plain: no API, no mobile app, no extension, and a blog announced but still empty.
What it does
- Search scientific literature across the preconnected arXiv corpus
- Ask questions directly to a paper in an interactive chat
- Generate an audio podcast from a paper and listen to it
- Upload your own PDF and work on it the same way
- Read the title, authors, full abstract and podcast transcript on a paper page
- Open the original PDF from the paper page
- Share a generated podcast through a public link
When to use PaperBrain / When not to
A quick filter to help you decide if PaperBrain is the right fit.
When to use PaperBrain
- Researchers, PhD students and postdocs who work through large volumes of arXiv literature every week
- Students who need to get inside a technical paper quickly, through its abstract, a chat and an audio version
- Auditory learners who would rather listen to a paper replayed as a spoken dialogue than read it on screen
- People working in arXiv-centric fields: physics, computer science and AI, mathematics, statistics, quantitative biology
- Individuals looking for a free, no-commitment tool they can use on their own account
When not to use PaperBrain
- Enterprise and public-sector buyers who need a contract and a published price: there are no terms of service, no DPA, no subprocessor list, and the pricing page requires a login
- Organisations operating under strict GDPR obligations: the privacy policy runs to two paragraphs and says nothing about retention or data subject rights
- Developers who want to embed the service in their own stack: there is no public API and no developer documentation
- Users who need a native mobile app: PaperBrain is web-only, with no iOS or Android release
- Readers whose literature sits outside arXiv, since the default corpus is arXiv and anything else has to be uploaded as a PDF
How to use PaperBrain
A typical end-to-end flow, from setup to results.
- Create an account from the 'Get Started' button on the homepage; sign-up and sign-in are handled by the Kinde authentication provider
- Search the arXiv corpus for the paper you need from the search page
- Or upload your own PDF from the upload page if the paper is not in the corpus
- Open the paper and read its title, authors and full abstract
- Put your questions to the paper in the interactive chat
- Generate an audio podcast from the paper and choose a register: funny, moderate or expert
- Listen to the podcast and follow the transcript displayed next to the paper
- Open the original PDF from the paper page if you want to check a passage
- Share the result with a public podcast link, which opens without an account
- Find your work again from your personal dashboard; subscription matters sit on the billing page
Pros & Cons
Pros
- Free to use with a simple account, with no time-limited trial counting down
- An audio format that is rare in this segment: the paper becomes a listenable dialogue
- The same paper can be rendered at several levels of difficulty, from playful to expert
- Podcast pages are public and shareable, and open without an account
- The arXiv corpus is already connected, so there is nothing to configure before the first search
- Your own PDFs can be processed in the same way as indexed papers
- A single web interface, with no installation and no technical prerequisite
Cons
- No public pricing: the pricing page redirects to the authentication screen, so no amount can be checked before signing up
- The terms of service page is published but empty, carrying a heading and no text at all
- The privacy policy is two paragraphs of boilerplate, and the sections it announces are simply missing
- No contact email on the domain: the only exposed contact is the founder's personal Gmail address
- No postal address and no named legal entity anywhere on the site
- No API, no mobile app, no extension, and a blog announced but still empty
- A single declared contributor, so continuity rests on one person, and most of the product can only be assessed after signing up
Pricing & Plans
PaperBrain offers a free plan. Its FAQ states plainly that the tool is free to use and that a free account is enough to start. A paid plan also exists, presented as a way to support the platform's development. No amount, no currency and no billing period is published anywhere: the pricing page redirects to the Kinde authentication screen, and no public archive capture of that page exists since the site went live in 2024. A billing route exists inside the application, also behind the login. The cost of the paid plan can therefore only be established after creating an account.
- the only plan named publicly
- described in the FAQ as enough to start using the platform
- with no detail on what it includes
- referred to only as 'a paid plan' supporting the platform's development
- with no commercial name
- no price and no feature list
- and no pricing table reachable without an account
Data, GDPR & hosting
A consolidated view of how PaperBrain handles your data.
GDPR overview
There is no GDPR compliance information of any kind. A search across every page collected, in both text and HTML, returns no occurrence of 'GDPR', 'RGPD' or 'CCPA'. No EU representative under Article 27 is designated, no data protection officer is named, and no privacy contact address is given. The privacy policy states only that 'It is PaperBrain's policy to respect your privacy regarding any information we may collect while operating our website'. It sets out no legal basis for processing, describes no right of access, rectification or erasure, mentions no international transfer mechanism or standard contractual clauses, and gives no postal address for a controller. This is a verified absence, not a gap in research: PaperBrain claims no GDPR compliance, and none can be inferred.
Who owns the data?
No document published by PaperBrain states who owns the content handled by the service. The privacy policy runs to two paragraphs and never addresses ownership. The terms of service page, which would normally carry such clauses, is published but empty apart from its heading. Nothing at all is said about uploaded PDFs: no ownership, no licence granted to the operator, no deletion. The only data the policy names are non-identifying browsing details (browser type, language preference, referring site, date and time of each request) and IP addresses of logged-in users and blog commenters. On ownership, PaperBrain makes no commitment in either direction.
Reuse rights
The privacy policy states a single purpose for the non-identifying information PaperBrain collects, namely browser type, language preference, referring site and the date and time of each request: understanding how visitors use the website. It adds that aggregated data may be published, for instance as reports on usage trends. IP addresses are collected for logged-in users and for people commenting on blog posts. The policy announces that disclosure to third parties is covered 'as described below', but no such section exists on the published page. Nothing is said about whether user content feeds model training, and nothing defines what an end user may reuse or republish: the terms of service page that would normally set those rules contains no text.
Data retention & training
Hosting summary
No hosting jurisdiction is declared. The privacy policy names no country and no region where data is stored, and the terms of service page, which would normally carry such a clause, contains no text. What can be established is technical rather than contractual: the domain resolves to 66.33.60.67, an address geolocated in the United States and announced by AS16509 (Amazon), which is an anycast node and therefore says little about where data actually rests. Podcast audio files are served from an Amazon S3 bucket named paperbrain-podcasts in the ap-south-1 region. These are observations made from the outside, not commitments by the publisher. PaperBrain undertakes nothing on hosting location, and an organisation with data residency requirements has no document it could rely on.
Where PaperBrain works
Country-level availability.
Not available in
Things to keep in mind
Risks and trade-offs to weigh before adopting PaperBrain.
- The terms of service page is published but empty: there is no contractual commitment you could hold the operator to, on anything
- The privacy policy is truncated. It announces a section on disclosure to third parties that does not exist on the published page
- No GDPR mention anywhere, no EU representative under Article 27, no data protection officer and no privacy contact address
- Nothing is said about how long data is kept, what happens to an uploaded PDF, or whether user content feeds model training: treat unpublished or confidential manuscripts as unsuitable for upload
- Prices are invisible before you create an account, which makes any budget decision impossible upfront
- The '100k+ Researchers' claim cannot be verified, and the logo strip mixes a university logo with generic images
- Listening to a generated podcast is not reading the paper: a dialogue rendered in a 'funny' or 'moderate' register compresses and reformulates, so any claim you intend to rely on should be checked against the original PDF
Setup & Integrations
Technical difficulty
Immediate. PaperBrain is a fully web-based application: there is nothing to install, no API key to obtain and no connector to configure. The only prerequisite is creating an account, handled by the Kinde authentication provider from the 'Get Started' button. The arXiv corpus is available from the first search, so a new user can find a paper, open a chat or generate a podcast within minutes. Uploading a PDF requires no preparation beyond having the file to hand. No technical skills are needed at any point.
Deployment
Integrations
Behind PaperBrain
Fundraising
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What is PaperBrain?
Is PaperBrain free?
Where do the papers come from?
What exactly is the podcast feature?
Do I need an account to use it?
How much does the paid plan cost?
Is there an API?
Is there a mobile app?
What does the privacy policy say?
Who publishes PaperBrain?
Should you pick PaperBrain?
PaperBrain is a real, working product rather than a landing page: its application routes are live, and podcast pages generated from genuine arXiv papers can be opened and listened to without an account. The differentiator is real too, turning a paper into a multi-voice audio dialogue and offering the same work at three registers, funny, moderate and expert. Entry is free, and that is the best argument for trying it: a free account, no trial clock running down, the arXiv corpus already connected, and the option to upload your own PDFs. For an individual researcher or student who reads a great deal, the proposition stands on its own. The reservations are serious, and they concern transparency rather than the product. The terms of service page exists but contains no text, so there is no contractual commitment to rely on. The privacy policy is two paragraphs and announces a section on third-party disclosure that was never written. Nothing is said about retention, deletion, or what becomes of an uploaded PDF. There is no GDPR statement, no EU representative, no data protection officer and no privacy contact. No price is published, and no legal entity or postal address is named. Add a solo project started in 2024, a copyright frozen at that year and an empty blog, and the picture is clear enough. PaperBrain is worth trying on a personal account, with material you would be comfortable seeing leave your machine. It is hard to recommend in a professional, contractual or regulated setting until its publisher issues real terms, a genuine privacy policy and a price.
- Choosing a selection results in a full page refresh.
- Opens in a new window.