PeopleofSuperintelligence

Researcher

David Wagner

Also known as David A. Wagner

David Wagner is a UC Berkeley computer scientist whose work tests how machine-learning defenses withstand attacks. He earned a mathematics degree at Princeton in 1995 and computer-science graduate degrees at Berkeley in 1999 and 2000; his early work addressed software and election security. With Anish Athalye and Nicholas Carlini, he showed how obfuscated gradients could make defenses against adversarial examples appear secure, and their 2018 paper won ICML's Best Paper Award. As of October 2026, he is a professor and Carl J. Penther Chair in Engineering at Berkeley, where he works on AI for cybersecurity and security for AI systems.

Works
347
Citations
31,450
h-index
65
Badges

Publication numbers from OpenAlex, as of 19 October 2026.

Timeline

  1. 1995

    A Princeton mathematics degree

    At Princeton, he earns an A.B. in mathematics in 1995.

  2. 1996

    Confining helper applications

    With Ian Goldberg, Randi Thomas, and Eric Brewer, he publishes A Secure Environment for Untrusted Helper Applications in 1996, proposing to reduce risk by restricting helper programs' access to the operating system.

  3. 1999–2000

    Computer science graduate degrees

    He completes a computer-science M.S. at UC Berkeley in 1999 and earns a Ph.D. in computer science there in 2000, advised by Eric Brewer.

  4. 2007

    Reviewing voting systems

    As a Berkeley associate professor, he leads the university team in California's 2007 review of three electronic voting systems. The researchers find common security problems in all three.

  5. 2018

    Exposing gradient masking

    With Anish Athalye and Nicholas Carlini, he publishes a 2018 paper that identifies obfuscated gradients as a false signal of security in adversarial-example defenses. Their attacks circumvent six defenses completely and one partially under the original threat models.

  6. 2018

    An ICML Best Paper

    ICML awards Best Paper to Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples, coauthored by Athalye, Carlini, and Wagner in 2018.

  7. 2019

    A USENIX Test of Time Award

    He and Eric Brewer, Ian Goldberg, and Randi Thomas receive the 2019 USENIX Test of Time Award for their 1996 paper A Secure Environment for Untrusted Helper Applications.

  8. As of October 2026

    Professor and Penther Chair

    As of October 2026, he is a professor and Carl J. Penther Chair in Engineering in UC Berkeley's Computer Science Division. He works on AI for cybersecurity and security for AI systems, including large language models.

Career

  1. Since 1996University of California, BerkeleyProfessor and Carl J. Penther Chair in Engineering
  2. 2007–2026University of New Brunswick

Education

  1. 2000University of California, BerkeleyPh.D. in computer science
  2. 1999University of California, BerkeleyM.S. in computer science
  3. 1995Princeton UniversityA.B. in mathematics

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. Towards Evaluating the Robustness of Neural Networks2017 · IEEE Symposium on Security and Privacy · 8,823 citationsWith Nicholas Carlini
  2. Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples2018 · 2,171 citationsWith Nicholas Carlini
  3. Adversarial Examples Are Not Easily Detected2017 · ACM Workshop on Artificial Intelligence and Security (AISec) · 1,344 citationsWith Nicholas Carlini
  4. Android permissions demystified2011 · Proceedings of the ACM Conference on Computer and Communications Security · 1,305 citationsWith Dawn Song
  5. Audio Adversarial Examples: Targeted Attacks on Speech-to-Text2018 · IEEE Security and Privacy Workshops (SPW) · 1,069 citationsWith Nicholas Carlini

From OpenAlex, as of 19 October 2026.

  1. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes University of California, Berkeley. Our estimates of where the people shaping AI studied and where their careers led, from education records in cited biographies.
  2. Google has the highest combined h-index among AI organizationsIncludes University of California, Berkeley. Our estimates of combined and median h-index at AI organizations, using current affiliations and OpenAlex records.
  3. The median h-index is 53 at universities and 19 at frontier AI labsIncludes University of California, Berkeley. The median h-index is 53 at universities and 19 at frontier AI labs. We compare publishing careers, citations and moves to estimate how universities and frontier labs contribute to AI research.
  4. 44% of AI career moves go from one university to anotherIncludes University of California, Berkeley. We estimate how 14,272 AI career moves link universities, companies, labs and government research centers, with net flows and the most common paths.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of David Wagner yet. Send us a free-licensed photo.

  1. David Wagner | EECS at UC Berkeley www2.eecs.berkeley.edu. Checked 19 October 2026.
  2. A Secure Environment for Untrusted Helper Applications | USENIX usenix.org. Checked 19 October 2026.
  3. David Wagner: debugging the vote - Berkeley Graduate Division grad.berkeley.edu. Checked 19 October 2026.
  4. Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples proceedings.mlr.press. Checked 19 October 2026.
  5. ICML 2018 Awards icml.cc. Checked 19 October 2026.
  6. David Wagner, Eric Brewer, Ian Goldberg, and Randi Thomas win 2019 USENIX Test of Time Award eecs.berkeley.edu. Checked 19 October 2026.
  7. David Wagner people.eecs.berkeley.edu. Checked 19 October 2026.
  8. David Wagner | Research UC Berkeley vcresearch.berkeley.edu. Checked 19 October 2026.