View source for MD5

{{Short description|Message-digest hashing algorithm}}
{{Use dmy dates|date=April 2014}}
{{Infobox cryptographic hash function
| name = MD5

| caption = 
<!-- General -->
| designers = [[Ronald Rivest]]
| publish date = April 1992
| series = [[MD2 (cryptography)|MD2]], [[MD4]], MD5, [[MD6]]
| derived from =
| derived to =
| related to =
| certification =
<!-- Detail  -->
| digest size = 128 bit
| block size = 512 bit
| structure = [[Merkle–Damgård construction]]
| rounds = 4<ref>{{cite IETF |title=The MD5 Message-Digest Algorithm |rfc=1321 |sectionname=Step 4. Process Message in 16-Word Blocks |section=3.4 |page=5 |last=Rivest |first=R. |author-link=Ron Rivest |date=April 1992 |publisher=[[Internet Engineering Task Force|IETF]] |access-date=2018-10-10 |doi=10.17487/RFC1321 }}</ref>
| cryptanalysis = A 2013 attack by Xie Tao, Fanbao Liu, and Dengguo Feng breaks MD5 [[collision resistance]] in 2<sup>18</sup> time. This attack runs in less than a second on a regular computer.<ref>{{Cite web |author1=Xie Tao |author2=Fanbao Liu |author3=Dengguo Feng |year=2013 |title=Fast Collision Attack on MD5 |website=Cryptology ePrint Archive |url=https://eprint.iacr.org/2013/170.pdf |access-date=3 December 2013 |archive-date=2 February 2021 |archive-url=https://web.archive.org/web/20210202021701/https://eprint.iacr.org/2013/170.pdf |url-status=live }}</ref>
MD5 is prone to [[length extension attack]]s.
}}{{Wikifunctions|Z10137}}

The '''MD5 message-digest algorithm''' is a widely used [[hash function]] producing a 128-[[bit]] hash value. MD5 was designed by [[Ronald Rivest]] in 1991 to replace an earlier hash function [[MD4]],<ref name="Ron Barak">{{cite book|last=Ciampa|first=Mark|title=CompTIA Security+ 2008 in depth|year=2009|publisher=Course Technology/Cengage Learning|location=Australia; United States|page=[https://archive.org/details/comptiasecurity20000ciam/page/290 290]|url=https://archive.org/details/comptiasecurity20000ciam|url-access=registration|isbn=978-1-59863-913-1}}</ref> and was specified in 1992 as [[Request for Comments|RFC]] 1321.

MD5 can be used as a [[checksum]] to verify [[data integrity]] against unintentional corruption. Historically it was widely used as a [[cryptographic hash function]]; however it has been found to suffer from extensive vulnerabilities. It remains suitable for other non-cryptographic purposes, for example for determining the partition for a particular key in a [[Partition (database)|partitioned database]], and may be preferred due to lower computational requirements than more recent [[Secure Hash Algorithms]].<ref>{{cite book |last1=Kleppmann |first1=Martin |title=Designing Data-Intensive Applications: The Big Ideas Behind Reliable, Scalable, and Maintainable Systems |date=April 2, 2017 |publisher=O'Reilly Media |isbn=978-1449373320 |page=203 |edition=1}}</ref>

==History and cryptanalysis==
MD5 is one in a series of [[message digest]] algorithms designed by Professor [[Ronald Rivest]] of [[Massachusetts Institute of Technology|MIT]] (Rivest, 1992).  When analytic work indicated that MD5's predecessor [[MD4]] was likely to be insecure, Rivest designed MD5 in 1991 as a secure replacement. ([[Hans Dobbertin]] did indeed later find weaknesses in MD4.)

In 1993, Den Boer and Bosselaers gave an early, although limited, result of finding a "[[hash collision|pseudo-collision]]" of the MD5 [[One-way compression function|compression function]]; that is, two different [[initialization vector]]s that produce an identical digest.

In 1996, Dobbertin announced a collision of the compression function of MD5 (Dobbertin, 1996). While this was not an attack on the full MD5 hash function, it was close enough for cryptographers to recommend switching to a replacement, such as [[SHA-1]] (also compromised since) or [[RIPEMD-160]].

The size of the hash value (128 bits) is small enough to contemplate a [[birthday attack]]. [[MD5CRK]] was a [[distributed computing|distributed project]] started in March 2004 to demonstrate that MD5 is practically insecure by finding a collision using a birthday attack.

MD5CRK ended shortly after 17 August 2004, when [[hash collision|collisions]] for the full MD5 were announced by [[Xiaoyun Wang]], Dengguo Feng, [[Xuejia Lai]], and Hongbo Yu.<ref name="autogenerated2" /><ref>{{cite journal |last1=Hawkes |first1=Philip |last2=Paddon |first2=Michael |last3=Rose |first3=Gregory G. |author-link3=Gregory G. Rose |title=Musings on the Wang et al. MD5 Collision |journal=[[Cryptology ePrint Archive]] |date=13 Oct 2004 |url=https://eprint.iacr.org/2004/264 |access-date=10 October 2018 |archive-url=https://web.archive.org/web/20181105220829/https://eprint.iacr.org/2004/264 |archive-date=5 November 2018 }}</ref> Their analytical attack was reported to take only one hour on an [[IBM p690]] cluster.<ref>{{cite web|url=http://www.bishopfox.com/resources/tools/other-free-tools/md4md5-collision-code/|title=Fast MD5 and MD4 Collision Generators |website=BishopFox |date=26 September 2013|author=Bishop Fox|access-date=10 February 2014|archive-date=26 April 2017|archive-url=https://web.archive.org/web/20170426171733/http://www.bishopfox.com/resources/tools/other-free-tools/md4md5-collision-code/}}</ref>

On 1 March 2005, [[Arjen Lenstra]], [[Wang Xiaoyun|Xiaoyun Wang]], and Benne de Weger demonstrated construction of two [[X.509]] certificates with different public keys and the same MD5 hash value, a demonstrably practical collision.<ref>{{cite journal |last1=Lenstra |first1=Arjen |last2=Wang |first2=Xiaoyun |last3=Weger |first3=Benne de |author-link1=Arjen Lenstra |author-link2=Xiaoyun Wang |title=Colliding X.509 Certificates |journal=[[Cryptology ePrint Archive]] |date=1 Mar 2005 |url=http://eprint.iacr.org/2005/067 |access-date=10 October 2018 |archive-date=23 May 2017 |archive-url=https://web.archive.org/web/20170523000238/http://eprint.iacr.org/2005/067 |url-status=live }}</ref> The construction included private keys for both public keys. A few days later, [[Vlastimil Klima]] described an improved algorithm, able to construct MD5 collisions in a few hours on a single notebook computer.<ref>{{cite journal |last1=Klíma |first1=Vlastimil |author-link1=Vlastimil Klíma |title=Finding MD5 Collisions&nbsp;– a Toy For a Notebook |journal=[[Cryptology ePrint Archive]] |date=5 Mar 2005 |url=http://eprint.iacr.org/2005/075 |access-date=10 October 2018 |archive-date=17 May 2017 |archive-url=https://web.archive.org/web/20170517175404/http://eprint.iacr.org/2005/075 |url-status=live }}</ref> On 18 March 2006, Klima published an algorithm that could find a collision within one minute on a single notebook computer, using a method he calls tunneling.<ref>Vlastimil Klima: [http://eprint.iacr.org/2006/105 Tunnels in Hash Functions: MD5 Collisions Within a Minute] {{Webarchive|url=https://web.archive.org/web/20110806052832/http://eprint.iacr.org/2006/105 |date=6 August 2011 }}, Cryptology ePrint Archive Report 2006/105, 18 March 2006, revised 17 April 2006. Retrieved 27 July 2008.</ref>

Various MD5-related [[Request for Comments#Obtaining RFCs|RFC errata]] have been published. 
In 2009, the [[United States Cyber Command]] used an MD5 hash value of their mission statement as a part of their official emblem.<ref>{{cite magazine
|url= https://www.wired.com/dangerroom/2010/07/code-cracked-cyber-command-logos-mystery-solved/
|title= Code Cracked! Cyber Command Logo Mystery Solved
|magazine= [[United States Cyber Command|USCYBERCOM]]
|publisher= [[Wired News]]
|date= 8 July 2010
|access-date= 29 July 2011
|archive-date= 17 February 2014
|archive-url= https://web.archive.org/web/20140217212938/http://www.wired.com/dangerroom/2010/07/code-cracked-cyber-command-logos-mystery-solved/
|url-status= live
}}</ref>

On 24 December 2010, Tao Xie and Dengguo Feng announced the first published single-block (512-bit) MD5 collision.<ref>{{cite web
|url=http://eprint.iacr.org/2010/643
|title=Construct MD5 Collisions Using Just A Single Block Of Message
|year=2010
|format=PDF
|author1=Tao Xie
|author2=Dengguo Feng
|access-date=28 July 2011
|archive-date=14 May 2017
|archive-url=https://web.archive.org/web/20170514185806/http://eprint.iacr.org/2010/643
|url-status=live
}}</ref> (Previous collision discoveries had relied on multi-block attacks.) For "security reasons", Xie and Feng did not disclose the new attack method. They issued a challenge to the cryptographic community, offering a US$10,000 reward to the first finder of a different 64-byte collision before 1 January 2013. [[Marc Stevens (cryptology)|Marc Stevens]] responded to the challenge and published colliding single-block messages as well as the construction algorithm and sources.<ref>{{cite web |url= http://marc-stevens.nl/research/md5-1block-collision/ |title= Marc Stevens – Research – Single-block collision attack on MD5 |publisher= Marc-stevens.nl |date= 2012 |access-date= 10 April 2014 |archive-date= 15 May 2017 |archive-url= https://web.archive.org/web/20170515070731/http://marc-stevens.nl/research/md5-1block-collision/ |url-status= live }}</ref>

In 2011 an informational RFC 6151<ref>{{cite journal|url=https://tools.ietf.org/html/rfc6151|title=RFC 6151 – Updated Security Considerations for the MD5 Message-Digest and the HMAC-MD5 Algorithms|website=Internet Engineering Task Force|date=March 2011|access-date=11 November 2013|last1=Turner|first1=Sean|doi=10.17487/RFC6151|archive-date=15 June 2017|archive-url=https://web.archive.org/web/20170615213134/https://tools.ietf.org/html///rfc6151|url-status=live}}</ref> was approved to update the security considerations in MD5<ref>{{cite journal |url=https://tools.ietf.org/html/rfc1321 |title=RFC 1321 – The MD5 Message-Digest Algorithm |website=Internet Engineering Task Force |date=April 1992 |access-date=5 October 2013 |last1=Rivest |first1=Ronald L. |doi=10.17487/RFC1321 |doi-access=free |archive-date=9 April 2021 |archive-url=https://web.archive.org/web/20210409200613/https://tools.ietf.org/html/rfc1321 |url-status=live |hdl=1721.1/149165 |hdl-access=free }}</ref> and HMAC-MD5.<ref>{{cite journal |url= https://tools.ietf.org/html/rfc2104 |title= RFC 2104 – HMAC: Keyed-Hashing for Message Authentication |website= Internet Engineering Task Force |date= February 1997 |access-date= 5 October 2013 |last1= Krawczyk |first1= Hugo |last2= Bellare |first2= Mihir |last3= Canetti |first3= Ran |doi= 10.17487/RFC2104 |archive-date= 15 April 2021 |archive-url= https://web.archive.org/web/20210415003434/https://tools.ietf.org/html/rfc2104 |url-status= live }}</ref>

==Security==
One basic requirement of any cryptographic hash function is that it should be [[computational complexity theory#Intractability|computationally infeasible]] to find two distinct messages that hash to the same value. MD5 fails this requirement catastrophically. On 31 December 2008, the [[CMU Software Engineering Institute]] concluded that  MD5 was essentially "cryptographically broken and unsuitable for further use".<ref name=Dougherty2008>{{cite web|first1=Chad R.|last1=Dougherty|title=Vulnerability Note VU#836068 MD5 vulnerable to collision attacks|url=https://www.kb.cert.org/vuls/id/836068|website=Vulnerability notes database|publisher=CERT Carnegie Mellon University Software Engineering Institute|access-date=3 February 2017|date=31 Dec 2008|archive-date=26 July 2011|archive-url=https://web.archive.org/web/20110726144513/http://www.kb.cert.org/vuls/id/836068|url-status=live}}</ref> The weaknesses of MD5 have been exploited in the field, most infamously by the [[Flame malware]] in 2012. {{As of|2019}}, MD5 continues to be widely used, despite its well-documented weaknesses and deprecation by security experts.<ref name=Cimpanu2019/>

A [[collision attack]] exists that can find [[collision resistance|collisions]] within seconds on a computer with a 2.6&nbsp;GHz Pentium 4 processor (complexity of 2<sup>24.1</sup>).<ref>{{Cite thesis |degree=Master's |author=M.M.J. Stevens |date=June 2007 |title=On Collisions for MD5 |url=http://www.win.tue.nl/hashclash/On%20Collisions%20for%20MD5%20-%20M.M.J.%20Stevens.pdf |access-date=31 March 2010 |archive-date=17 May 2017 |archive-url=https://web.archive.org/web/20170517115509/http://www.win.tue.nl/hashclash/On%20Collisions%20for%20MD5%20-%20M.M.J.%20Stevens.pdf |url-status=live }}</ref> Further, there is also a [[chosen-prefix collision attack]] that can produce a collision for two inputs with specified prefixes within seconds, using off-the-shelf computing hardware (complexity 2<sup>39</sup>).<ref>{{Cite web |author1=Marc Stevens |author2=Arjen Lenstra |author3=Benne de Weger |date=16 June 2009 |title=Chosen-prefix Collisions for MD5 and Applications |website=École Polytechnique Fédérale de Lausanne |url=https://documents.epfl.ch/users/l/le/lenstra/public/papers/lat.pdf |access-date=31 March 2010 |archive-url=https://web.archive.org/web/20111109092157/https://documents.epfl.ch/users/l/le/lenstra/public/papers/lat.pdf |archive-date=9 November 2011 }}</ref>
The ability to find collisions has been greatly aided by the use of off-the-shelf [[Graphics processing unit|GPUs]]. On an NVIDIA GeForce 8400GS graphics processor, 16–18 million hashes per second can be computed. An NVIDIA GeForce 8800 Ultra can calculate more than 200 million hashes per second.<ref>{{cite web
 | url = http://bvernoux.free.fr/md5/index.php
 | title = New GPU MD5 cracker cracks more than 200 million hashes per second.
 | access-date = 25 March 2011
 | archive-date = 11 May 2011
 | archive-url = https://web.archive.org/web/20110511121239/http://bvernoux.free.fr/md5/index.php
 | url-status = live
 }}</ref>

These hash and collision attacks have been demonstrated in the public in various situations, including colliding document files<ref>{{cite web |author=Magnus Daum, [[Stefan Lucks]] |title=Hash Collisions (The Poisoned Message Attack) |work=[[Eurocrypt]] 2005 rump session |url=http://th.informatik.uni-mannheim.de/People/lucks/HashCollisions/ |archive-url=https://web.archive.org/web/20100327141611/http://th.informatik.uni-mannheim.de/people/lucks/HashCollisions/ |archive-date=27 March 2010 |df=dmy-all }}</ref><ref name=special-file-formats>{{Cite web |author1=Max Gebhardt |author2=Georg Illies |author3=Werner Schindler |title=A Note on the Practical Value of Single Hash Collisions for Special File Formats |website=National Institute of Standards and Technology |url=http://csrc.nist.gov/groups/ST/hash/documents/Illies_NIST_05.pdf |archive-url=https://web.archive.org/web/20080917182949/http://csrc.nist.gov/groups/ST/hash/documents/Illies_NIST_05.pdf |archive-date=2008-09-17 |date=31 October 2005}}</ref> and [[digital certificate]]s.<ref name="sslHarmful" /> As of 2015, MD5 was demonstrated to be still quite widely used, most notably by security research and antivirus companies.<ref>{{Cite web|title = Poisonous MD5 – Wolves Among the Sheep {{!}} Silent Signal Techblog|date = 10 June 2015|url = http://blog.silentsignal.eu/2015/06/10/poisonous-md5-wolves-among-the-sheep/|access-date = 2015-06-10|archive-date = 10 June 2015|archive-url = https://web.archive.org/web/20150610115424/http://blog.silentsignal.eu/2015/06/10/poisonous-md5-wolves-among-the-sheep/|url-status = live}}</ref>

As of 2019, one quarter of widely used [[content management system]]s were reported to still use MD5 for [[password hashing]].<ref name=Cimpanu2019>{{Cite web|url=https://www.zdnet.com/article/a-quarter-of-major-cmss-use-outdated-md5-as-the-default-password-hashing-scheme/|title=A quarter of major CMSs use outdated MD5 as the default password hashing scheme|last=Cimpanu|first=Catalin|website=ZDNet|language=en|access-date=2019-06-17|archive-date=24 January 2021|archive-url=https://web.archive.org/web/20210124103125/https://www.zdnet.com/article/a-quarter-of-major-cmss-use-outdated-md5-as-the-default-password-hashing-scheme/|url-status=live}}</ref>

=== Overview of security issues ===
In 1996, a flaw was found in the design of MD5. While it was not deemed a fatal weakness at the time, cryptographers began recommending the use of other algorithms, such as [[SHA-1]], which has since been found to be vulnerable as well.<ref>{{cite web|url=http://ftp.arnes.si/packages/crypto-tools/rsa.com/cryptobytes/crypto2n2.pdf.gz|title=The Status of MD5 After a Recent Attack|author=Hans Dobbertin|work=CryptoBytes|volume=2|issue=2|date=Summer 1996|access-date=22 October 2013}}</ref>
In 2004 it was shown that MD5 is not [[collision-resistant]].<ref>{{cite web|url=http://merlot.usc.edu/csac-f06/papers/Wang05a.pdf|title=How to Break MD5 and Other Hash Functions|author1=Xiaoyun Wang|author2=Hongbo Yu |work=Advances in Cryptology – Lecture Notes in Computer Science|volume=3494|pages=19–35|year=2005|access-date=21 December 2009|archive-url=https://web.archive.org/web/20090521024709/http://merlot.usc.edu/csac-f06/papers/Wang05a.pdf|archive-date=21 May 2009}}</ref> As such, MD5 is not suitable for applications like [[Transport Layer Security|SSL]] [[public key certificate|certificates]] or [[digital signature]]s that rely on this property for digital security. Researchers additionally discovered more serious flaws in MD5, and described a feasible [[collision attack]]—a method to create a pair of inputs for which MD5 produces identical [[checksum]]s.<ref name="autogenerated2">J. Black, M. Cochran, T. Highland: [http://www.cs.colorado.edu/~jrblack/papers/md5e-full.pdf A Study of the MD5 Attacks: Insights and Improvements] {{Webarchive|url=https://web.archive.org/web/20150101093005/http://www.cs.colorado.edu/%7Ejrblack/papers/md5e-full.pdf |date=1 January 2015 }}, 3 March 2006. Retrieved 27 July 2008.</ref><ref name="autogenerated1">Xiaoyun Wang, Dengguo ,k.,m.,m, HAVAL-128 and [[RIPEMD]], Cryptology ePrint Archive Report 2004/199, 16 August 2004, revised 17 August 2004. Retrieved 27 July 2008.</ref> Further advances were made in breaking MD5 in 2005, 2006, and 2007.<ref>Marc Stevens, Arjen Lenstra, Benne de Weger: [http://www.win.tue.nl/hashclash/SoftIntCodeSign/ Vulnerability of software integrity and code signing applications to chosen-prefix collisions for MD5] {{Webarchive|url=https://web.archive.org/web/20071213023720/http://www.win.tue.nl/hashclash/SoftIntCodeSign/ |date=13 December 2007 }}, 30 November 2007. Retrieved 27 July 2008.</ref> In December 2008, a group of researchers used this technique to fake [[SSL certificate]] validity.<ref name="sslHarmful">{{cite web |url=http://www.win.tue.nl/hashclash/rogue-ca/ |title=MD5 considered harmful today |last=Sotirov |first=Alexander |author2=Marc Stevens |author3=Jacob Appelbaum |author4=Arjen Lenstra |author5=David Molnar |author6=Dag Arne Osvik |author7=Benne de Weger |date=30 December 2008 |access-date=30 December 2008 |archive-date=25 March 2017 |archive-url=https://web.archive.org/web/20170325033522/http://www.win.tue.nl/hashclash/rogue-ca/ |url-status=live }} [https://events.ccc.de/congress/2008/Fahrplan/events/3023.en.html Announced] {{Webarchive|url=https://web.archive.org/web/20181116081156/https://events.ccc.de/congress/2008/Fahrplan/events/3023.en.html |date=16 November 2018 }} at the 25th [[Chaos Communication Congress]].</ref><ref name="browserflaw">{{cite web
|url=http://news.cnet.com/8301-1009_3-10129693-83.html
|title=Web browser flaw could put e-commerce security at risk
|last=Stray
|first=Jonathan
|date=30 December 2008
|access-date=24 February 2009
|publisher=[[CNET.com]]
|archive-date=28 August 2013
|archive-url=https://web.archive.org/web/20130828142658/http://news.cnet.com/8301-1009_3-10129693-83.html
}}</ref>

As of 2010, the [[CMU Software Engineering Institute]] considers MD5 "cryptographically broken and unsuitable for further use",<ref name=Dougherty2008/> and most U.S. government applications now require the [[SHA-2]] family of hash functions.<ref>{{cite web |url=http://csrc.nist.gov/groups/ST/hash/policy.html |title=NIST.gov&nbsp;— Computer Security Division&nbsp;— Computer Security Resource Center |publisher=Csrc.nist.gov |access-date=9 August 2010 |archive-url=https://web.archive.org/web/20110609064344/http://csrc.nist.gov/groups/ST/hash/policy.html |archive-date=9 June 2011 }}</ref>  In 2012, the [[Flame (malware)|Flame]] malware exploited the weaknesses in MD5 to fake a Microsoft [[digital signature]].<ref name="foo">{{cite web|url=http://blogs.technet.com/b/srd/archive/2012/06/06/more-information-about-the-digital-certificates-used-to-sign-the-flame-malware.aspx|title=Flame malware collision attack explained|access-date=7 June 2012|archive-url=https://web.archive.org/web/20120608225029/http://blogs.technet.com/b/srd/archive/2012/06/06/more-information-about-the-digital-certificates-used-to-sign-the-flame-malware.aspx|archive-date=8 June 2012}}</ref>

===Collision vulnerabilities===
{{Further|Collision attack}}

In 1996, collisions were found in the compression function of MD5, and [[Hans Dobbertin]] wrote in the [[RSA Laboratories]] technical newsletter, "The presented attack does not yet threaten practical applications of MD5, but it comes rather close ... in the future MD5 should no longer be implemented ... where a collision-resistant hash function is required."<ref>{{Cite FTP |url=ftp://ftp.rsasecurity.com/pub/cryptobytes/crypto2n2.pdf
|date=Summer 1996
|volume=2
|issue=2
|page=1
|title=The Status of MD5 After a Recent Attack
|last=Dobbertin
|first=Hans
|access-date=10 August 2010
|server=RSA Laboratories CryptoBytes
|url-status=dead
|quote=The presented attack does not yet threaten practical applications of MD5, but it comes rather close. ....{{sic}} in the future MD5 should no longer be implemented...{{sic}} where a collision-resistant hash function is required.
}}</ref>

In 2005, researchers were able to create pairs of [[PostScript]] documents<ref>{{cite web |url=http://www.schneier.com/blog/archives/2005/06/more_md5_collis.html |title=Schneier on Security: More MD5 Collisions |publisher=Schneier.com |access-date=9 August 2010 |archive-date=11 April 2021 |archive-url=https://web.archive.org/web/20210411035935/https://www.schneier.com/blog/archives/2005/06/more_md5_collis.html |url-status=live }}</ref> and [[X.509]] certificates<ref>{{cite web |url=http://www.win.tue.nl/~bdeweger/CollidingCertificates/ |title=Colliding X.509 Certificates |publisher=Win.tue.nl |access-date=9 August 2010 |archive-date=15 May 2017 |archive-url=https://web.archive.org/web/20170515022608/http://www.win.tue.nl/~bdeweger/CollidingCertificates/ |url-status=live }}</ref> with the same hash. Later that year, MD5's designer Ron Rivest wrote that "md5 and sha1 are both clearly broken (in terms of collision-resistance)".<ref>{{cite web |url=http://mail.python.org/pipermail/python-dev/2005-December/058850.html |title=[Python-Dev&#93; hashlib&nbsp;— faster md5/sha, adds sha256/512 support |date=16 December 2005 |publisher=Mail.python.org |access-date=9 August 2010 |archive-date=6 May 2021 |archive-url=https://web.archive.org/web/20210506232819/https://mail.python.org/pipermail/python-dev/2005-December/058850.html |url-status=live }}</ref>

On 30 December 2008, a group of researchers announced at the 25th [[Chaos Communication Congress]] how they had used MD5 collisions to create an intermediate certificate authority certificate that appeared to be legitimate when checked by its MD5 hash.<ref name="sslHarmful" /> The researchers used a [[PS3 cluster]] at the [[École Polytechnique Fédérale de Lausanne|EPFL]] in [[Lausanne]], Switzerland<ref>{{cite magazine|url=http://blog.wired.com/27bstroke6/2008/12/berlin.html|title=Researchers Use PlayStation Cluster to Forge a Web Skeleton Key|date=31 December 2008|magazine=Wired|access-date=31 December 2008|archive-date=21 April 2009|archive-url=https://web.archive.org/web/20090421023048/http://blog.wired.com/27bstroke6/2008/12/berlin.html|url-status=live}}</ref> to change a normal SSL certificate issued by [[RapidSSL]] into a working [[CA certificate]] for that issuer, which could then be used to create other certificates that would appear to be legitimate and issued by RapidSSL. [[Verisign]], the issuers of RapidSSL certificates, said they stopped issuing new certificates using MD5 as their checksum algorithm for RapidSSL once the vulnerability was announced.<ref>{{cite web|url=https://blogs.verisign.com/ssl-blog/2008/12/on_md5_vulnerabilities_and_mit.php|title=This morning's MD5 attack&nbsp;— resolved|last=Callan|first=Tim|date=31 December 2008|publisher=Verisign|access-date=31 December 2008|archive-url=https://web.archive.org/web/20090116180944/http://blogs.verisign.com/ssl-blog/2008/12/on_md5_vulnerabilities_and_mit.php|archive-date=16 January 2009}}</ref> Although Verisign declined to revoke existing certificates signed using MD5, their response was considered adequate by the authors of the exploit ([[Alexander Sotirov]], [[Marc Stevens (Cryptology)|Marc Stevens]], [[Jacob Appelbaum]], [[Arjen Lenstra]], David Molnar, Dag Arne Osvik, and Benne de Weger).<ref name="sslHarmful" /> Bruce Schneier wrote of the attack that "we already knew that MD5 is a broken hash function" and that "no one should be using MD5 anymore".<ref>{{cite web |author=Bruce Schneier |url=http://www.schneier.com/blog/archives/2008/12/forging_ssl_cer.html |title=Forging SSL Certificates |publisher=Schneier on Security |date=31 December 2008 |access-date=10 April 2014 |archive-date=9 November 2020 |archive-url=https://web.archive.org/web/20201109014745/https://www.schneier.com/blog/archives/2008/12/forging_ssl_cer.html |url-status=live }}</ref> The SSL researchers wrote, "Our desired impact is that Certification Authorities will stop using MD5 in issuing new certificates. We also hope that use of MD5 in other applications will be reconsidered as well."<ref name="sslHarmful" />

In 2012, according to [[Microsoft]], the authors of the [[Flame (malware)|Flame]] malware used an MD5 collision to forge a Windows code-signing certificate.<ref name="foo" />

MD5 uses the [[Merkle–Damgård construction]], so if two prefixes with the same hash can be constructed, a common suffix can be added to both to make the collision more likely to be accepted as valid data by the application using it. Furthermore, current collision-finding techniques allow specifying an arbitrary ''prefix'': an attacker can create two colliding files that both begin with the same content. All the attacker needs to generate two colliding files is a template file with a 128-byte block of data, aligned on a 64-byte boundary, that can be changed freely by the collision-finding algorithm. An example MD5 collision, with the two messages differing in 6 bytes, is:

 d131dd02c5e6eec4 693d9a0698aff95c 2fcab5{{Background color|#87CEEB|8}}712467eab 4004583eb8fb7f89
 55ad340609f4b302 83e4888325{{Background color|#87CEEB|7}}1415a 085125e8f7cdc99f d91dbd{{Background color|#87CEEB|f}}280373c5b
 d8823e3156348f5b ae6dacd436c919c6 dd53e2{{Background color|#87CEEB|b}}487da03fd 02396306d248cda0
 e99f33420f577ee8 ce54b67080{{Background color|#87CEEB|a}}80d1e c69821bcb6a88393 96f965{{Background color|#87CEEB|2}}b6ff72a70

 d131dd02c5e6eec4 693d9a0698aff95c 2fcab5{{Background color|#87CEEB|0}}712467eab 4004583eb8fb7f89
 55ad340609f4b302 83e4888325{{Background color|#87CEEB|f}}1415a 085125e8f7cdc99f d91dbd{{Background color|#87CEEB|7}}280373c5b
 d8823e3156348f5b ae6dacd436c919c6 dd53e2{{Background color|#87CEEB|3}}487da03fd 02396306d248cda0
 e99f33420f577ee8 ce54b67080{{Background color|#87CEEB|2}}80d1e c69821bcb6a88393 96f965{{Background color|#87CEEB|a}}b6ff72a70

Both produce the MD5 hash <code>79054025255fb1a26e4bc422aef54eb4</code>.<ref>{{cite web
|url=http://www.rtfm.com/movabletype/archives/2004_08.html#001055
|title=A real MD5 collision
|author=Eric Rescorla
|date=2004-08-17
|archive-url=https://web.archive.org/web/20140815234704/http://www.rtfm.com/movabletype/archives/2004_08.html#001055
|archive-date=2014-08-15
|work=Educated Guesswork (blog)
|access-date=2015-04-13}}</ref>
The difference between the two samples is that the leading bit in each [[nibble]] has been flipped. For example, the 20th byte (offset 0x13) in the top sample, 0x87, is 10000111 in binary. The leading bit in the byte (also the leading bit in the first nibble) is flipped to make 00000111, which is 0x07, as shown in the lower sample.

Later it was also found to be possible to construct collisions between two files with separately chosen prefixes. This technique was used in the creation of the rogue CA certificate in 2008. A new variant of parallelized collision searching using [[Message Passing Interface|MPI]] was proposed by Anton Kuznetsov in 2014, which allowed finding a collision in 11 hours on a computing cluster.<ref>{{cite web | url=http://eprint.iacr.org/2014/871.pdf | title=An algorithm for MD5 single-block collision attack using high performance computing cluster | publisher=IACR | access-date=2014-11-03 | author=Anton A. Kuznetsov | archive-date=4 June 2016 | archive-url=https://web.archive.org/web/20160604093753/https://eprint.iacr.org/2014/871.pdf | url-status=live }}</ref>

===Preimage vulnerability===
In April 2009, an attack against MD5 was published that breaks MD5's [[preimage resistance]]. This attack is only theoretical, with a computational complexity of 2<sup>123.4</sup> for full preimage.<ref>{{Cite book|author1=Yu Sasaki |title=Advances in Cryptology - EUROCRYPT 2009 |volume=5479 |pages=134–152 |author2=Kazumaro Aoki |date=16 April 2009 |chapter=Finding Preimages in Full MD5 Faster Than Exhaustive Search |publisher=[[Springer Berlin Heidelberg]] |doi=10.1007/978-3-642-01001-9_8 |series=Lecture Notes in Computer Science |isbn=978-3-642-01000-2 }}</ref><ref>{{cite book
|chapter=Construction of the Initial Structure for Preimage Attack of MD5
|publisher=[[Institute of Electrical and Electronics Engineers|IEEE]] Computer Society
|year=2009|volume=1|pages=442–445|author=Ming Mao and Shaohui Chen and Jin Xu
|title=2009 International Conference on Computational Intelligence and Security
|doi=10.1109/CIS.2009.214|isbn=978-0-7695-3931-7
|s2cid=16512325
}}</ref>

==Applications==
MD5 digests have been widely used in the [[software]] world to provide some assurance that a transferred file has arrived intact. For example, file servers often provide a pre-computed MD5 (known as [[md5sum]]) [[checksum]] for the files, so that a user can compare the checksum of the downloaded file to it. Most unix-based operating systems include MD5 sum utilities in their distribution packages; Windows users may use the included [[PowerShell]] function "Get-FileHash", the included command line function "certutil -hashfile <filename> md5",<ref>{{cite web |title=Finding Checksum Values in Windows 10 |url=https://answers.microsoft.com/en-us/windows/forum/all/finding-checksum-values-in-windows-10/dbc3c569-4b5a-4967-8810-c25255cdc1fd |publisher=Microsoft Community |access-date=23 November 2023 |archive-date=11 January 2024 |archive-url=https://web.archive.org/web/20240111185213/https://answers.microsoft.com/en-us/windows/forum/all/finding-checksum-values-in-windows-10/dbc3c569-4b5a-4967-8810-c25255cdc1fd |url-status=live }}</ref><ref>{{cite web |title=certutil |url=https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/certutil |website=certutil |publisher=Microsoft Learn |access-date=23 November 2023 |archive-date=23 November 2023 |archive-url=https://web.archive.org/web/20231123190431/https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/certutil |url-status=live }}</ref> install a Microsoft utility,<ref>{{cite web |url=https://support.microsoft.com/kb/841290/en-us |title=Availability and description of the File Checksum Integrity Verifier utility |publisher=Microsoft Support |date=17 June 2013 |access-date=10 April 2014 |archive-date=15 February 2015 |archive-url=https://web.archive.org/web/20150215153720/http://support.microsoft.com/kb/841290/en-us |url-status=live }}</ref><ref>{{cite web |url=https://support.microsoft.com/kb/889768/en-us |title=How to compute the MD5 or SHA-1 cryptographic hash values for a file |publisher=Microsoft Support |date=23 January 2007 |access-date=10 April 2014 |archive-date=9 March 2015 |archive-url=https://web.archive.org/web/20150309230754/http://support.microsoft.com/kb/889768/en-us |url-status=live }}</ref> or use third-party applications. Android ROMs also use this type of checksum.

[[File:CPT-Hashing-File-Transmission.svg|350px|center|Diagram showing use of MD5 hashing in file transmission]]

As it is easy to generate MD5 collisions, it is possible for the person who created the file to create a second file with the same checksum, so this technique cannot protect against some forms of malicious tampering. In some cases, the checksum cannot be trusted (for example, if it was obtained over the same channel as the downloaded file), in which case MD5 can only provide error-checking functionality: it will recognize a corrupt or incomplete download, which becomes more likely when downloading larger files.

Historically, MD5 has been used to store a one-way hash of a [[Password#Form of stored passwords|password]], often with [[key stretching]].<ref>{{Cite web|url = https://www.freebsd.org/cgi/man.cgi?crypt(3)|title = FreeBSD Handbook, Security – DES, Blowfish, MD5, and Crypt|access-date = 2014-10-19|archive-date = 18 February 2017|archive-url = https://web.archive.org/web/20170218081024/https://www.freebsd.org/cgi/man.cgi?crypt(3)|url-status = live}}</ref><ref>{{cite web |url=http://docs.oracle.com/cd/E26505_01/html/816-5174/policy.conf-4.html |title=Synopsis – man pages section 4: File Formats |publisher=Docs.oracle.com |date=1 January 2013 |access-date=10 April 2014 |archive-date=4 March 2016 |archive-url=https://web.archive.org/web/20160304204917/http://docs.oracle.com/cd/E26505_01/html/816-5174/policy.conf-4.html |url-status=live }}</ref> [[NIST]] does not include MD5 in their list of recommended hashes for password storage.<ref>[http://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-132.pdf NIST SP 800-132] {{Webarchive|url=https://web.archive.org/web/20161201044854/http://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-132.pdf |date=1 December 2016 }} Section 5.1</ref>

MD5 is also used in the field of [[electronic discovery]], to provide a unique identifier for each document that is exchanged during the legal discovery process. This method can be used to replace the [[Bates numbering|Bates stamp]] numbering system that has been used for decades during the exchange of paper documents. As above, this usage should be discouraged due to the ease of collision attacks.

==Algorithm==
[[Image:MD5 algorithm.svg|right|thumbnail|300px|Figure 1. One MD5 operation. MD5 consists of 64 of these operations, grouped in four rounds of 16 operations. {{mvar|F}} is a nonlinear function; one function is used in each round. {{math|''M''<sub>''i''</sub>}} denotes a 32-bit block of the message input, and {{math|''K''<sub>''i''</sub>}} denotes a 32-bit constant, different for each operation. {{math|&lt;&lt;&lt;<sub>''s''</sub>}} denotes a left bit rotation by {{mvar|s}} places; {{mvar|s}} varies for each operation. <math>\boxplus</math> denotes addition modulo 2<sup>32</sup>.]]

MD5 processes a variable-length message into a fixed-length output of 128 bits. The input message is broken up into chunks of 512-bit blocks (sixteen 32-bit words); the message is always [[padding (cryptography)|padded]] even if its original length is divisible by 512 (see RFC 1321, section 3.1). The padding works as follows: first, a single bit, 1, is appended to the end of the message. This is followed by as many zeros as are required to bring the length of the message up to 64 bits fewer than a multiple of 512. The remaining bits are filled up with 64 bits representing the length of the original message, modulo 2<sup>64</sup>.

The main MD5 algorithm operates on a 128-bit state, divided into four 32-bit words, denoted {{mvar|A}}, {{mvar|B}}, {{mvar|C}}, and {{mvar|D}}. These are initialized to certain fixed constants. The main algorithm then uses each 512-bit message block in turn to modify the state. The processing of a message block consists of four similar stages, termed ''rounds''; each round is composed of 16 similar operations based on a non-linear function {{mvar|F}}, modular addition, and left rotation. Figure 1 illustrates one operation within a round. There are four possible functions; a different one is used in each round:
:<math>\begin{align}
F(B,C,D) &= (B\wedge{C}) \vee (\neg{B} \wedge{D}) \\
G(B,C,D) &= (B\wedge{D}) \vee (C \wedge \neg{D}) \\
H(B,C,D) &= B \oplus C \oplus D \\
I(B,C,D) &= C \oplus (B \vee \neg{D})
\end{align}
</math>

<math>\oplus, \wedge, \vee, \neg</math> denote the [[XOR]], [[Logical conjunction|AND]], [[Logical disjunction|OR]] and [[Negation|NOT]] operations respectively.

===Pseudocode===
The MD5 hash is calculated according to this algorithm.<ref>{{Cite web|url=https://referencesource.microsoft.com/#System.Workflow.Runtime/MD5HashHelper.cs,5a97802b6014fccc,references|title=Reference Source|access-date=23 December 2020|archive-date=21 June 2021|archive-url=https://web.archive.org/web/20210621192842/https://referencesource.microsoft.com/#System.Workflow.Runtime/MD5HashHelper.cs,5a97802b6014fccc,references|url-status=live}}</ref> All values are in [[Endianness|little-endian]].

 <span style="color:green;">// '': All variables are unsigned 32 bit and wrap modulo 2^32 when calculating''</span>
 '''var''' ''int'' s[64], K[64]
 '''var''' ''int'' i
 
 <span style="color:green;">// ''s specifies the per-round shift amounts''</span>
 s[ 0..15] := { 7, 12, 17, 22,  7, 12, 17, 22,  7, 12, 17, 22,  7, 12, 17, 22 }
 s[16..31] := { 5,  9, 14, 20,  5,  9, 14, 20,  5,  9, 14, 20,  5,  9, 14, 20 }
 s[32..47] := { 4, 11, 16, 23,  4, 11, 16, 23,  4, 11, 16, 23,  4, 11, 16, 23 }
 s[48..63] := { 6, 10, 15, 21,  6, 10, 15, 21,  6, 10, 15, 21,  6, 10, 15, 21 }
 
 <span style="color:green;">// ''Use binary integer part of the sines of integers (Radians) as constants:''</span>
 '''for''' i '''from''' 0 '''to''' 63 '''do'''
     K[i] := floor(2<sup>32</sup> × abs(sin(i + 1)))
 '''end for'''
 <span style="color:green;">// ''(Or just use the following precomputed table):''</span>
 K[ 0.. 3] := { 0xd76aa478, 0xe8c7b756, 0x242070db, 0xc1bdceee }
 K[ 4.. 7] := { 0xf57c0faf, 0x4787c62a, 0xa8304613, 0xfd469501 }
 K[ 8..11] := { 0x698098d8, 0x8b44f7af, 0xffff5bb1, 0x895cd7be }
 K[12..15] := { 0x6b901122, 0xfd987193, 0xa679438e, 0x49b40821 }
 K[16..19] := { 0xf61e2562, 0xc040b340, 0x265e5a51, 0xe9b6c7aa }
 K[20..23] := { 0xd62f105d, 0x02441453, 0xd8a1e681, 0xe7d3fbc8 }
 K[24..27] := { 0x21e1cde6, 0xc33707d6, 0xf4d50d87, 0x455a14ed }
 K[28..31] := { 0xa9e3e905, 0xfcefa3f8, 0x676f02d9, 0x8d2a4c8a }
 K[32..35] := { 0xfffa3942, 0x8771f681, 0x6d9d6122, 0xfde5380c }
 K[36..39] := { 0xa4beea44, 0x4bdecfa9, 0xf6bb4b60, 0xbebfbc70 }
 K[40..43] := { 0x289b7ec6, 0xeaa127fa, 0xd4ef3085, 0x04881d05 }
 K[44..47] := { 0xd9d4d039, 0xe6db99e5, 0x1fa27cf8, 0xc4ac5665 }
 K[48..51] := { 0xf4292244, 0x432aff97, 0xab9423a7, 0xfc93a039 }
 K[52..55] := { 0x655b59c3, 0x8f0ccc92, 0xffeff47d, 0x85845dd1 }
 K[56..59] := { 0x6fa87e4f, 0xfe2ce6e0, 0xa3014314, 0x4e0811a1 }
 K[60..63] := { 0xf7537e82, 0xbd3af235, 0x2ad7d2bb, 0xeb86d391 }
 
 <span style="color:green;">// ''Initialize variables:''</span>
 '''var''' ''int'' a0 := 0x67452301   <span style="color:green;">// A</span>
 '''var''' ''int'' b0 := 0xefcdab89   <span style="color:green;">// B</span>
 '''var''' ''int'' c0 := 0x98badcfe   <span style="color:green;">// C</span>
 '''var''' ''int'' d0 := 0x10325476   <span style="color:green;">// D</span>
 
 <span style="color:green;">// ''Pre-processing: adding a single 1 bit''</span>
 '''append''' "1" bit '''to''' message<    
  // Notice: the input bytes are considered as bit strings,
  //  where the first bit is the most significant bit of the byte.<ref>RFC 1321, section 2, "Terminology and Notation", Page 2.</ref>
 
 <span style="color:green;">// ''Pre-processing: padding with zeros''</span>
 '''append''' "0" bit '''until''' message length in bits ≡ 448 (mod 512)
 
 <span style="color:green;">// Notice: the two padding steps above are implemented in a simpler way
   //  in implementations that only work with complete bytes: append 0x80
   //  and pad with 0x00 bytes so that the message length in bytes ≡ 56 (mod 64).</span>
 
 '''append''' original length in bits '''mod''' 2<sup>64</sup> '''to''' message
 
 <span style="color:green;">// ''Process the message in successive 512-bit chunks:''</span>
 '''for each''' ''512-bit'' chunk '''of''' padded message '''do'''
     break chunk into sixteen 32-bit words M[j], 0 ≤ j ≤ 15
 <span style="color:green;">    // ''Initialize hash value for this chunk:''</span>
     '''var''' ''int'' A := a0
     '''var''' ''int'' B := b0
     '''var''' ''int'' C := c0
     '''var''' ''int'' D := d0
 <span style="color:green;">    // ''Main loop:''</span>
     '''for''' i '''from''' 0 '''to''' 63 '''do'''
         '''var''' ''int'' F, g
         '''if''' 0 ≤ i ≤ 15 '''then'''
             F := (B '''and''' C) '''or''' (('''not''' B) '''and''' D)
             g := i
         '''else if''' 16 ≤ i ≤ 31 '''then'''
             F := (D '''and''' B) '''or''' (('''not''' D) '''and''' C)
             g := (5×i + 1) '''mod''' 16
         '''else if''' 32 ≤ i ≤ 47 '''then'''
             F := B '''xor''' C '''xor''' D
             g := (3×i + 5) '''mod''' 16
         '''else if''' 48 ≤ i ≤ 63 '''then'''
             F := C '''xor''' (B '''or''' ('''not''' D))
             g := (7×i) '''mod''' 16
 <span style="color:green;">        // ''Be wary of the below definitions of a,b,c,d''</span>
         F := F + A + K[i] + M[g] <span style="color:green;"> // ''M[g] must be a 32-bit block''</span>
         A := D
         D := C
         C := B
         B := B + '''leftrotate'''(F, s[i])
     '''end for'''
 <span style="color:green;">    // ''Add this chunk's hash to result so far:''</span>
     a0 := a0 + A
     b0 := b0 + B
     c0 := c0 + C
     d0 := d0 + D
 '''end for'''
 
 '''var''' ''char'' digest[16] := a0 '''append''' b0 '''append''' c0 '''append''' d0 <span style="color:green;">// ''(Output is in little-endian)''</span>

Instead of the formulation from the original RFC 1321 shown, the following may be used for improved efficiency (useful if assembly language is being used – otherwise, the compiler will generally optimize the above code. Since each computation is dependent on another in these formulations, this is often slower than the above method where the nand/and can be parallelised):
 ( 0 ≤ i ≤ 15): F := D '''xor''' (B '''and''' (C '''xor''' D))
 (16 ≤ i ≤ 31): F := C '''xor''' (D '''and''' (B '''xor''' C))

== MD5 hashes ==
The 128-bit (16-byte) MD5 hashes (also termed ''message digests'') are typically represented as a sequence of 32 [[hexadecimal]] digits. The following demonstrates a 43-byte [[ASCII]] input and the corresponding MD5 hash:

 MD5("[[The quick brown fox jumps over the lazy dog]]") =
 9e107d9d372bb6826bd81d3542a419d6

Even a small change in the message will (with overwhelming probability) result in a mostly different hash, due to the [[avalanche effect]]. For example, adding a period to the end of the sentence:

 MD5("[[The quick brown fox jumps over the lazy dog]]{{Background color|#87CEEB|'''.'''}}") = 
 e4d909c290d0fb1ca068ffaddf22cbd0

The hash of the zero-length string is:

 MD5("") = 
 d41d8cd98f00b204e9800998ecf8427e

The MD5 algorithm is specified for messages consisting of any number of bits; it is not limited to multiples of eight bits ([[Octet (computing)|octets]], [[byte]]s). Some MD5 implementations such as [[md5sum]] might be limited to octets, or they might not support ''streaming'' for messages of an initially undetermined length.

== Implementations ==
Below is a list of cryptography libraries that support MD5:

* [[Botan (programming library)|Botan]]
* [[Bouncy Castle (cryptography)|Bouncy Castle]]
* [[cryptlib]]
* [[Crypto++]]
* [[Libgcrypt]]
* [[Nettle (cryptographic library)|Nettle]]
* [[OpenSSL]]
* [[wolfSSL]]

==See also==
* [[Comparison of cryptographic hash functions]]
* [[Hash function security summary]]
* [[HashClash]]
* [[Crypt (C)#MD5-based scheme|MD5Crypt]]
* [[md5deep]]
* [[md5sum]]
* [[MD6]]
* [[SHA-1]]
* [[SHA-2]]

==References==
{{Reflist|30em}}

== Further reading ==
* {{Cite conference
 | first = Thomas A.
 | last = Berson
 | title = Differential Cryptanalysis Mod 2<sup>32</sup> with Applications to MD5
 | book-title = EUROCRYPT
 | year = 1992
 | pages = 71–80
 | isbn = 3-540-56413-6
}}
* {{Cite conference
 |author1=Bert den Boer |author2=Antoon Bosselaers | title = Collisions for the Compression Function of MD5
 | conference = EUROCRYPT
 | book-title = Advances in Cryptology – EUROCRYPT '93
 | year = 1993
 | pages = 293–304
 | isbn = 978-3-540-57600-6
 | publisher = Springer
 | location = Berlin; London
}}
* Hans Dobbertin, Cryptanalysis of MD5 compress. Announcement on Internet, May 1996. {{cite web |url=http://citeseer.ist.psu.edu/dobbertin96cryptanalysis.html |title=CiteSeerX |publisher=Citeseer.ist.psu.edu |access-date=9 August 2010 |archive-date=24 June 2008 |archive-url=https://web.archive.org/web/20080624043846/http://citeseer.ist.psu.edu/dobbertin96cryptanalysis.html |url-status=live }}
* {{Cite journal
 | first = Hans
 | last = Dobbertin
 | title = The Status of MD5 After a Recent Attack
 | journal = CryptoBytes
 | volume = 2
 | issue = 2
 | year = 1996
 | url = http://ftp.arnes.si/packages/crypto-tools/rsa.com/cryptobytes/crypto2n2.pdf.gz
}}
* {{Cite conference
 | author1 = Xiaoyun Wang
 | author2 = Hongbo Yu
 | title = How to Break MD5 and Other Hash Functions
 | book-title = EUROCRYPT
 | year = 2005
 | url = http://www.infosec.sdu.edu.cn/uploadfile/papers/How%20to%20Break%20MD5%20and%20Other%20Hash%20Functions.pdf
 | isbn = 3-540-25910-4
 | access-date = 6 March 2008
 | archive-url = https://web.archive.org/web/20090521001714/http://www.infosec.sdu.edu.cn/uploadfile/papers/How%20to%20Break%20MD5%20and%20Other%20Hash%20Functions.pdf
 | archive-date = 21 May 2009
 }}

==External links==
<!--======================== {{No more links}} ============================
 | PLEASE BE CAUTIOUS IN ADDING MORE LINKS TO THIS ARTICLE. Wikipedia |
 | is not a collection of links nor should it be used for advertising. |
 | |
 | Excessive or inappropriate links WILL BE DELETED. |
 | See [[Wikipedia:External links]] & [[Wikipedia:Spam]] for details. |
 | |
 | If there are already plentiful links, please propose additions or |
 | replacements on this article's discussion page, or submit your link |
 | to the relevant category at the Open Directory Project (dmoz.org) |
 | and link back to that category using the {{dmoz}} template. |
 ======================= {{No more links}} =============================-->
* [http://www.w3.org/TR/1998/REC-DSig-label/MD5-1_0 W3C recommendation on MD5] {{Webarchive|url=https://web.archive.org/web/20141228122551/http://www.w3.org/TR/1998/REC-DSig-label/MD5-1_0 |date=28 December 2014 }}
* [https://fe-tool.com/en-us/hash/md5 MD5 Calculator] {{Webarchive|url=https://web.archive.org/web/20221116025344/https://fe-tool.com/en-us/hash/md5 |date=16 November 2022 }}

{{Cryptography navbox|hash}}

[[Category:Articles with example pseudocode]]
[[Category:Broken hash functions]]