Internet Archive

• 1. Introduction

  ›

  The Internet Archive holds more than one trillion web pages, a number its founder Brewster Kahle called a civilization-scale milestone when it was reached in October 2025. That single figure raises a question worth sitting with: how does a non-profit library founded in 1996 become the custodian of a civilization's digital memory? The Archive does not simply store snapshots of dead websites. It preserves software programs, 78 rpm phonograph cylinders, dissertations written before living memory, concert recordings, government films, and the entire television coverage of September 11 as it aired in real time. What follows is the story of how that ambition was built, attacked, sued, and defended.

• 2. Kahle's Founding Vision

  ›

  Brewster Kahle launched the Archive in May 1996, the same month he started the commercial web-crawling company Alexa Internet. The two ventures were not contradictions. Alexa would generate revenue; the Archive would preserve what Alexa's crawlers found. Kahle's stated mission was "universal access to all knowledge", a phrase broad enough to sound like a slogan and specific enough to generate decades of legal conflict.

  The first page ever saved was a December 1996 edition of USA Today. By October of that year the Archive was already sweeping the public web in large volumes. The archived content did not become easily accessible to the general public until 2001, when the Wayback Machine launched as a public interface.

  Kahle's vision grew outward from the web almost immediately. By the end of 1999 the Archive had expanded into physical media, beginning with the Prelinger Archives, a collection of ephemeral film. Audio, moving images, and software followed. Kahle articulated his ambition for physical books by invoking the Svalbard Global Seed Vault: he envisioned collecting one copy of every book ever published. He acknowledged the impossibility with a plainness that became characteristic. "We're not going to get there," he said, "but that's our goal."

  One telling window into Kahle's values is the story of Aaron Swartz. Kahle revealed in 2013 that Swartz had coordinated the bulk download of public-domain books from Google Book Search, distributing the task across enough computers and a slow enough pace to stay within Google's restrictions. The goal was to ensure those books reached the public domain permanently. Kahle described this as an example of Swartz's "genius" for working on what could give the most to the public good. The Archive's ceramic collection at its San Francisco headquarters reflects a similar instinct: more than one hundred sculpted figures represent Archive employees, and the one-hundredth figure immortalizes Swartz.

• 3. The Wayback Machine

  ›

  A joint effort between Alexa Internet and the Archive, the Wayback Machine began as a tool for seeing what websites used to look like and evolved into an infrastructure for preserving sites that no longer exist at all. By 2025 it held more than 866 billion web pages alongside 42.5 million print materials, 13 million videos, and 14 million audio files.

  The trillion-webpage threshold, announced at an event in San Francisco on the 22nd of October 2025, amounted to more than 100,000 terabytes of data. The milestone arrived in the same year that the U.S. Senate designated the Internet Archive as a Federal Depository Library, on the 24th of July 2025, granting it the right to store public access government records.

  Google's relationship with the Wayback Machine shifted notably in September 2024. Google announced it would include links to the Wayback Machine in its "more about this page" menu in Google Search, effectively absorbing a function that Google's own Google Cache service had performed before Google retired it earlier that year. The collaboration marked a rare moment in which a major commercial platform publicly deferred to a non-profit archive for the function of historical web lookup.

  Archive-It, the Archive's subscription web service created in late 2005, extended these capabilities to outside institutions. By 2021 it was partnering with more than 800 universities, archives, libraries, and museums. In 2022 it was used to back up Ukrainian websites as the conflict there accelerated. By 2024 the Archive claimed more than 1,200 partners.

• 4. Books, Lawsuits, and Lending

  ›

  The Archive operates 33 scanning centers in five countries and once digitized approximately 1,000 books a day. As of 2025 its text collection exceeded 47 million items, and the American libraries sub-collection alone had grown to 3.9 million items.

  The scale of the book collection made conflict with publishers nearly inevitable. In June 2020, during the COVID-19 pandemic, the Archive launched the National Emergency Library, removing the one-book-per-physical-copy waiting restriction for 1.4 million digitized titles. Four large publishers, Hachette Book Group, Penguin Random House, HarperCollins, and John Wiley, filed suit before the United States District Court for the Southern District of New York on the 1st of June 2020.

  Judge Koeltl ruled against the Archive on the 24th of March 2023. The negotiated judgment of the 11th of August 2023 barred the Archive from lending books that had electronic copies on sale. The United States Court of Appeals for the Second Circuit upheld the ruling on the 4th of September 2024, calling the Archive's fair-use argument "unpersuasive". More than 500,000 books were taken down as a result.

  The ruling was not a complete defeat. The Second Circuit clarified that the restrictions did not apply to out-of-print books. For in-print books, the restriction only engaged if an electronic version was commercially available. The Archive retained the ability to offer short passages, including links keyed to specific page numbers for citations. A donation of 400,000 uncatalogued dissertations from Leiden University Library, covering work from 1851 to 2004, arrived in October 2024 and illustrated the appetite for expansion that the lawsuits had not extinguished. The collection included original theses by Niels Bohr, Marie Curie, Albert Einstein, and Carl Jung, among others.

• 5. Sound, Film, and Ceramic Figures

  ›

  Boston Public Library donated hundreds of thousands of 78 rpm discs to the Archive in 2017. That donation helped seed the Great 78 Project, which aims to digitize phonograph recordings from the period between 1880 and 1960. By September 2025 the project had digitized 400,000 recordings, developed in collaboration with the Archive of Contemporary Music and George Blood Audio.

  The music industry's response arrived in August 2023. Universal Music Group, Sony Music, and Concord, together with labels including Capitol Records and Arista Records, sued the Archive for $621 million in alleged damages over the Great 78 Project. The suit noted that the Music Modernization Act, passed by Congress in 2018, had extended copyright protection to pre-1972 recordings through 2067. In its defense, the Archive argued that the primitive sound quality of the original recordings qualified for fair-use preservation, that downloads were minimal, and that more than 95% of the collection was not available anywhere else. Both parties filed to drop the case in September 2025 after reaching undisclosed settlement terms.

  The Live Music Archive documents a parallel and less litigated tradition. It holds more than 170,000 concert recordings from independent artists and bands with permissive recording policies, including the Grateful Dead. Jordan Zevon allowed the Archive to host his father Warren Zevon's concert recordings, a collection spanning 1976 to 2001 that contains 126 concerts and 1,137 individual songs.

  The Archive's Visual Arts Residency, organized by Amir Saber Esfahani, brings artists into direct contact with the collections. Resident artists from the 2018 cohort included Taravat Talepasand and Jenny Odell. The residency has run continuously since 2018.

• 6. The 2024 Cyberattacks

  ›

  During the week of the 27th of May 2024, a group calling itself SN_BLACKMETA launched distributed denial-of-service attacks against the Archive, taking its services offline intermittently for several days. The group had possible links to Anonymous Sudan. The incident drew comparison to the 2023 British Library cyberattack, which affected the UK Web Archive.

  A second wave began on the 9th of October 2024. Archivist Jason Scott and security researcher Scott Helme confirmed DDoS attacks, site defacement, and a data breach. A pop-up on the defaced site read: "Have you ever felt like the Internet Archive runs on sticks and is constantly on the verge of suffering a catastrophic security breach? It just happened. See 31 million of you on HIBP!" The attackers stole email addresses and Bcrypt-hashed passwords from a file dated the 28th of September 2024, affecting approximately 31 million user accounts.

  On the 11th of October, Kahle stated that the data was safe and that service would return "in days, not weeks". The Wayback Machine came back in read-only form on the 13th of October. Two days later the main site remained largely offline because the team was "prioritizing keeping data safe at the expense of service availability".

  On the 20th of October, a separate group of threat actors stole unrotated API tokens and breached the Archive's Zendesk email support platform. This group claimed responsibility for the data breaches while clarifying that SN_BLACKMETA had been behind only the DDoS attacks. Reporting by Bleeping Computer posited that the attackers had targeted the Archive not for money but to gain standing in the data-trafficking community. By the 25th of October, full service was restored. The Archive's announcement that it was building a Canadian backup copy in November 2016 had anticipated exactly this kind of threat; that project took on new meaning after the breach.

• 7. Infrastructure and Redundancy

  ›

  The Archive is headquartered at 300 Funston Avenue in San Francisco, a former Christian Science Church that became its home in 2009. Before that, from 1996 to 2009, it operated from the Presidio of San Francisco, a former U.S. military base.

  As of 2019 the Archive had an annual budget of $37 million, drawing from web-crawling service revenue, partnerships, grants, donations, and the Kahle-Austin Foundation. A December 2019 fundraising campaign set a target of $6 million in donations. The Archive uses Ubuntu as its operating system for website servers.

  Data redundancy is built into the architecture. The Archive holds copies at the Bibliotheca Alexandrina in Egypt and at a facility in Amsterdam, as well as at data centers in San Francisco, Redwood City, and Richmond, California. By 2025 the Archive operated six data centers, with smaller facilities in other U.S. states, Canada, and Europe. All data centers conform to ISO/IEC 27001 information security standards. Since 2020, content has also been stored on the Filecoin decentralized network; by October 2023 one petabyte of data had been uploaded there.

  Trent University donated 250,000 books in 2018. Marygrove College's entire library collection arrived after the college closed in 2020. The Archive was officially designated a library by the state of California in 2007, and joined the International Internet Preservation Consortium. In 2025 it opened a new headquarters for its European branch on the 19th of September, the same year its Federal Depository Library status was confirmed by the U.S. Senate.