At this time, the Creative Commons (CC) Technology team has concluded that
generative AI development tooling does not pass a cost/benefit analysis for
contributing to our open source projects. Until further notice, we will not
accept CC Open Source submissions that include code or content generated with
AI.
Overhyped
Generative AI productivity claims are often championed (and paid for) by those
who reap a material benefit from adoption of AI tooling. Adoption rates are
inflated by a lack of choice when AI tools are included by default and often
cannot be disabled (EditorDavid, 2025). We’ve observed that arguments for their
adoption too often appeal to a fear of missing out or potential future
benefits.
Inefficiency and slop
Our current evaluation of generative AI development tooling is that it does not
offer a benefit to our open source projects nor to the user of the AI tooling.
Indeed, "when developers use AI tools, they take 19% longer than without—AI
makes them slower." (METR, 2025). A recent report from MIT NANDA indicates
"that 95% of organizations are getting zero return [from generative AI]."
(Challapally et al., 2025). AI tooling enables the creation of spurious work
for reviewers. For example, see the following comment on an AI slop
submission:
This code does not call curl. This is not a "POC" of anything than [sic]
suggesting you did this with an AI and that you do not understand what you're
doing here.
Please identify the exact line of curl source code that has the problem
you say exists.
(badger, 2025)
Counterproductive
Especially for our contributors, when we participate in work programs like
Google Summer of Code (GSoC) or Outreachy, AI tools train the wrong skills. We
want contributors who are learning the technologies that underpin our open
source projects.
I listened to an industry representative say that he wants students to have
“design thinking and the ability to ask questions” without recognizing that
the student use of AI tools he advocates will stunt their critical thinking
skills, making it impossible for them to ask questions and do interesting
“design thinking.”.
(Valeries, 2025)
Responsibility and solidarity
In addition to concerns with how AI development tools might negatively impact
our day-to-day, the Technology team has concerns about how generative AI
technology is created and implemented/deployed. Even if generative AI
development tools were actually helpful, the Technology team would be reluctant
to use tools that were developed in a way that harmed others, that are funded
by deployments/implementations that harm others, and that offer short-term
benefits at an existential cost.
References
badger. (2025, September 16). Comment on Stack Buffer Overflow in cURL Cookie
Parsing Leads to RCE. curl.
https://hackerone.com/reports/3340109#activity-36894924
Challapally, A., Pease, C., Raskar, R., Chari, P., (2025, July). The GenAI
Divide: STATE OF AI IN BUSINESS 2025. MIT NANDA.
https://www.artificialintelligence-news.com/wp-content/uploads/2025/08/ai_report_2025.pdf
EditorDavid. (2025, May 31). GitHub Users Angry at the Prospect of AI-Written
Issues From Copilot. Slashdot.
https://developers.slashdot.org/story/25/06/01/0049240/github-users-angry-at-the-prospect-of-ai-written-issues-from-copilot
METR. (2025, July 10). Measuring the Impact of Early-2025 AI on Experienced
Open-Source Developer Productivity. METR.
https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study/
Valeries, Barr. (2025, August 29). Feeling Cranky About AI and CS Education.
Communications of the ACM.
https://cacm.acm.org/blogcacm/feeling-cranky-about-ai-and-cs-education/