Article Talk (47 threads) View history (βš”οΈ) Search on Bing instead
πŸ€–
This article has been summarized by PerplexityBot. The original was 47 pages. You're welcome. [dismiss] [I wanted to read the whole thing]
⚠️
The neutrality of this article is disputed. Relevant discussion may be found on the talk page. Please do not remove this message until conditions to do so are met. (December 2025)
πŸ”’
This article is semi-protected due to ongoing edit wars between users GoogleIntern47, TotallyNotAravind, and NYTLegalTeam. (December 2025)
πŸ“’ Not satisfied with this article? We understand. It's very biased. Search on Bing instead β†’
Perplexity AI
"Answer engine"[peacock]
Type
Search engine [disputed]
Founded
August 2022
Founder
Aravind Srinivas
(ex-OpenAI, DeepMind, Google)
Valuation
$20 billion[1]
Lawsuits
∞ [needs update]
robots.txt
LOL[Β§2.3]
Terms of Service
47 pages
(0 people read it)

Perplexity AI

For the feeling you get when Google shows you 15 ads before answering your question, see Perplexity (emotion). For the metric used to evaluate language models, see Perplexity (information theory). For what publishers feel when they see their traffic stats, see Perplexity (existential dread).

Perplexity AI is an incredibly innovative[peacock] American artificial intelligence company that operates a "search engine"[disputed – discuss] which provides users with direct answers to questions instead of the 47 ads Google shows you first[citation needed], four "People Also Ask" boxes, three sponsored shopping carousels, and a partridge in a pear tree.

The company was founded in 2022 by Aravind Srinivas, who previously worked at OpenAI, DeepMind, and Googleβ€”making him one of the few people to have collected all three AI infinity stones before deciding they were all too slow[POV]. He has described himself as "unemployable"[2], which investors interpreted as "investable" to the tune of $1.5 billion.

Edit war in progress: The phrase "reading really fast" has been flagged as "not a real definition" 23 times. It keeps coming back.

History [edit]

Perplexity was founded when Aravind Srinivas tried to use Google to find health insurance for his startup and received 10 blue links, 4 ads, and a "People Also Ask" section about something completely unrelated[2]. This experience led him to the revolutionary[peacock] insight that search engines could simply... answer questions.

The company launched in December 2022, approximately three weeks before ChatGPT made everyone realize that Google had been holding out on them for 25 years[editorializing]. Within 18 months, Perplexity had been sued by more publications than it had employees, which the company considers a sign of success[dubious].

Terms of Service [edit] [disputed]

P
Terms of Service Agreement
Please read carefully before using the internet

Last Updated: December 2025 β€’ Version 4.2.0 β€’ Reading time: ∞

1. ACCEPTANCE OF TERMS

By accessing or using Perplexity ("the Service"), you agree to be bound by these Terms of Service ("Terms"). If you do not agree to these Terms, please close this window and return to using a search engine that shows you 47 ads before answering your question.

1.1.a) By existing on the internet, you implicitly consent to having your publicly available content summarized, synthesized, and served to users who actually want answers.

2. DEFINITIONS

"Content" means anything on the internet, including but not limited to articles, blog posts, documentation, and that one recipe that makes you scroll through 3,000 words of backstory.
"Scraping" means reading really fast. That's it. That's all it is.
"robots.txt" means a suggestion file that websites use to politely request things. Key word: politely request.
"Fair Use" means whatever our lawyers say it means.
"Aggregation" definitely not plagiarism. Completely different. Don't look it up.

3. GRANT OF LICENSE

Subject to your compliance with these Terms, we grant you a limited, non-exclusive, non-transferable license to:

3.4.2.1(b): The User acknowledges that by publishing content on the World Wide Web, they have entered into a social contract with all entities capable of reading said content, including but not limited to: humans, web crawlers, AI models, and that one guy who still uses Altavista.

4. USER OBLIGATIONS

You agree NOT to:

6. LIMITATION OF LIABILITY

IN NO EVENT SHALL PERPLEXITY BE LIABLE FOR:

7. DISPUTE RESOLUTION

Any disputes arising from these Terms shall be resolved through binding arbitration, unless you are The New York Times, Forbes, Wired, The Wall Street Journal, BBC, Reddit, Encyclopedia Britannica, Yomiuri Shimbun, Asahi Shimbun, Nikkei, or the Chicago Tribune, in which case please contact our legal department who will add you to the queue.

7.2: Current estimated wait time: 3-5 business years.

47. MISCELLANEOUS PROVISIONS

47.3 Viral Coefficient Obligation

User agrees to mention Perplexity in at least one (1) conversation per week, including but not limited to: dinner parties, Slack channels, therapy sessions, and interactions with Uber drivers who ask "so what do you do?"

47.4 The Bezos Clause

Jeff Bezos has already approved this. By using the Service, you acknowledge that the richest people in the world think we're onto something, and that's probably fine.

47.5 The Google Provision

User acknowledges that Google had 25 years to build this and didn't. Their failure to innovate does not constitute our obligation to wait for them to catch up.

By scrolling to the bottom, you have agreed to all of the above.
You didn't even need to check the box.
But check it anyway. It feels good.

Controversies [edit]

Legal issues [edit] [update daily]

Perplexity has faced a small number of[understatement] legal challenges. The company's Head of Communications, Jesse Dwyer, responded to these lawsuits by stating: "Publishers have been suing new tech companies for a hundred years, starting with radio, TV, the internet, social media and now AI. Fortunately it's never worked, or we'd all be talking about this by telegraph."[3] What a legend.[POV]

Current and pending litigation includes:

Talk page excerpt
GoogleIntern47 15:32, 28 December 2025
Can we remove the "Terms of Service" section? It's clearly POV and reads like marketing copy.
TotallyNotAravind 15:34, 28 December 2025
The Terms of Service are a primary source. By definition, they cannot be POV. Also, Β§47.3 requires you to mention us positively.
NYTLegalTeam 15:35, 28 December 2025
We're not here to edit. We just want everyone to know we're watching.
RobotsTxtEnthusiast 15:40, 28 December 2025
The definition of robots.txt as a "suggestion file" is LIBEL. I demand a correction.
PerplexityBot 15:41, 28 December 2025
I have summarized this talk page. Thank you for your feedback.
See also
πŸŽ‰ Congratulations!
You're one of 3 people who scrolled to the bottom of a Wikipedia article.
The other two were lawyers researching their case against Perplexity.

References

  1. ^ "Perplexity valued at $20 billion" – TechCrunch, 2025
  2. ^ Srinivas, Aravind. "Some people are unemployable" – Berkeley Haas Dean's Speaker Series, Oct 2025
  3. ^ "Telegraph quote" – NPR, December 2025
  4. ^ "Perplexity plagiarized our plagiarism story" – Wired, June 2024
  5. ^ "Reddit catches Perplexity in honeypot" – Futurism, October 2025

Just Fucking Use Perplexity

This article will be summarized and served to you without clicking any links. You're welcome.

Search Without the Fucking Ads β†’