PeopleofSuperintelligence

Researcher

Huiwen Chang

Huiwen Chang is an AI researcher who builds image-generation models, from the MaskGIT and Muse architectures at Google Research to the image generation in OpenAI's GPT-4o. After a computer science degree in the Yao class at Tsinghua University, she completed a Princeton PhD on personal photo enhancement in 2018, advised by Adam Finkelstein. As a research scientist at Google, she invented MaskGIT, a 2022 model that generates all image tokens at once and refines them, and Muse, a text-to-image Transformer introduced in 2023. She joined OpenAI in 2023, co-created GPT-4o's image generation, and in June 2025 was named among the new hires of Meta Superintelligence Labs. Today she works at Meta's TBD Lab.

Timeline

  1. 2015

    Palette-based photo recoloring

    After a B.S. in computer science from the Yao class at Tsinghua University's IIIS, she pursues a PhD at Princeton, advised by Adam Finkelstein. In July 2015, she publishes Palette-based Photo Recoloring at SIGGRAPH with Finkelstein, Ohad Fried, Yiming Liu and Stephen DiVerdi, an interactive tool that lets non-experts recolor an image by editing a color palette.

  2. 2015–2018

    A Princeton PhD in photo enhancement

    She interns at Adobe in Seattle in 2015 and receives a Microsoft fellowship in 2016. In 2018, she completes her Princeton computer science PhD with a dissertation titled Personal Photo Enhancement, advised by Adam Finkelstein.

  3. 2022

    MaskGIT at Google Research

    As a research scientist at Google Research, she invents MaskGIT, an image-generation architecture. In February 2022, she and Han Zhang, Lu Jiang, Ce Liu and William T. Freeman describe it as a bidirectional transformer that generates all image tokens at once and refines them, speeding up autoregressive decoding by up to 64 times.

  4. January 2023

    The Muse text-to-image model

    At Google Research in January 2023, she and 11 coauthors, including Han Zhang, introduce Muse, a text-to-image Transformer. Given a text embedding from a pretrained large language model, Muse learns to predict randomly masked image tokens, and its 3-billion-parameter version reaches an FID of 7.88 on zero-shot COCO evaluation.

  5. 2023–2025

    GPT-4o image generation at OpenAI

    She joins OpenAI in 2023 after more than four years as a research scientist at Google. At OpenAI, she co-creates GPT-4o's image generation, and she is among the authors of the GPT-4o System Card in October 2024.

  6. June 2025

    Joining Meta Superintelligence Labs

    In June 2025, Mark Zuckerberg names her in an internal memo introducing the new hires of Meta Superintelligence Labs, many of whom come from OpenAI, Anthropic and Google.

  7. As of October 2026

    Researcher at Meta's TBD Lab

    As of October 2026, she works at Meta's TBD Lab, with research interests in multimodal intelligence, generation and interaction.

  1. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI from 1961 to 2025, with a year-by-year chart.
  2. The United States employs an estimated 53% of people in our data with a known employment countryOur estimates of where AI careers are concentrated now and how countries’ shares of dated positions changed over time.
  3. 46% of living people in our data with known birth and employment countries work outside their birth countryOur estimate of cross-border careers in AI, from country of birth and the country of each current organization.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionOur estimates of where the people shaping AI studied and where their careers led, from education records in cited biographies.

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 Huiwen Chang yet. Send us a free-licensed photo.

  1. Huiwen Chang – Rising Stars in EECS 2017 risingstars2017.stanford.edu. Checked 19 October 2026.
  2. Palette-based Photo Recoloring gfx.cs.princeton.edu. Checked 19 October 2026.
  3. Congratulations to the Computer Science Department Class of 2018! cs.princeton.edu. Checked 19 October 2026.
  4. Zuckerberg's Meta Superintelligence Labs poaches top AI talent in Silicon Valley e.vnexpress.net. Checked 19 October 2026.
  5. MaskGIT: Masked Generative Image Transformer arxiv.org. Checked 19 October 2026.
  6. Muse: Text-To-Image Generation via Masked Generative Transformers arxiv.org. Checked 19 October 2026.
  7. Here Is Everyone Mark Zuckerberg Has Hired So Far for Meta’s ‘Superintelligence’ Team wired.com. Checked 19 October 2026.
  8. GPT-4o System Card arxiv.org. Checked 19 October 2026.
  9. Meta founds Superintelligence Labs with top acquisitions from OpenAI and Google the-decoder.com. Checked 19 October 2026.
  10. Huiwen Chang - Google Scholar scholar.google.com. Checked 19 October 2026.