AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

Buying for a business?Offer from Amazon

Get business pricing on tech for your team

  • Business-only prices and quantity discounts
  • Tax-exempt purchasing
  • Multiple users, one account, clear invoices
As an affiliate, we earn on qualifying purchases.

QGIS merged a proposal on Oct. 7 that prohibits AI-, LLM- and agent-assisted code contributions to the qgis/qgis repository. The policy replaces an earlier approach that allowed AI assistance subject to review and disclosure; the proposal’s author said the project may consider limited exceptions later.

The QGIS project has adopted a policy prohibiting AI-, LLM- and agent-assisted code contributions to its main repository, replacing an earlier policy that allowed such help under review and disclosure rules. The proposal, QEP-428, was merged into the project’s enhancement-proposals repository on Oct. 7, setting a clear restriction for contributors to qgis/qgis.

The proposal, titled “No AI contributions permitted,” was authored by QGIS core contributor Nyall Dawson as pull request 387. Its text says that contributions to the qgis/qgis repository must not use an LLM, AI tool or agent-based workflow. It also warns that if such use is discovered later, the contribution may be reverted and the contributor may face sanctions affecting their rights in the project.

QEP-428 passed its discussion period on Oct. 5, moved to a vote and was merged on Oct. 7, according to the pull request and its associated diff. The restriction addresses code contributions to the named repository; the source material does not establish that it applies to every QGIS-related project, discussion or use of AI outside that contribution process.

Dawson said in the pull-request discussion that the project’s earlier rules were not being followed in practice. He described the new approach as starting with a closed door, with possible exceptions to be considered individually. Examples he raised included using AI as an autocomplete substitute and using it for translations. The proposal also states that enforcement can include both code reversion and contributor-rights sanctions.

At a glance
reportWhen: Merged Oct. 7, 2026; surfaced in a wide…
The developmentQGIS merged QEP-428, a proposal banning AI-assisted code contributions to the project’s main repository.

A New Rule for QGIS Contributions

The decision changes the terms under which contributors can submit code to a major open-source geographic information system. Rather than asking maintainers to review AI-assisted work and requiring contributors to disclose that assistance, the new policy prohibits AI involvement in code contributions outright. That gives maintainers a more direct rule to apply, but also places responsibility on contributors to understand what counts as AI- or agent-assisted work.

The policy may also shape how other open-source projects weigh concerns about code review, attribution and contributor participation. In the discussion, Dawson said the previous approach had not worked as intended and that the project had already lost at least one “extremely high value” contributor as a result. That is Dawson’s account; the pull request does not identify the contributor or independently document the circumstances.

For users, the immediate development is a change in how contributions to QGIS’s main code repository are governed, not an announcement of a change to the software itself. The proposal does not report a technical feature release or describe a specific effect on existing QGIS installations.

Amazon

QGIS GIS software

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

From Disclosure to Prohibition

QEP-428 replaces the project’s earlier QEP-408 policy, which adapted a human-in-the-loop approach from LLVM’s AI Tool Use Policy. That earlier framework required contributors to review all LLM output, disclose AI assistance through “Assisted-by” trailers and avoid posting unreviewed agent-generated comments. Its premise was that AI could be used if people remained responsible for the output and were transparent about the assistance.

The new proposal takes a different position: it bars AI-assisted code contributions rather than setting conditions for acceptable use. Dawson’s discussion comments frame this as an initial position that might later be narrowed or adjusted through case-by-case decisions. The proposal’s merge records that QEP-428 was added to the enhancement proposals; the binding language is the policy text included in the proposal.

The change drew broader attention after a Hacker News submission dated Oct. 12. That visibility came several days after the merge, so the online discussion should not be mistaken for the date the project adopted the proposal.

“No LLM, AI or agent assisted code contributions are allowed in the qgis/qgis repository.”

— QEP-428, the merged proposal

Amazon

open-source GIS tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Scope and Enforcement Questions

The proposal’s text names the qgis/qgis repository, but the available source material does not clarify how maintainers will interpret every kind of AI assistance, including tools that provide autocomplete or other limited suggestions. Dawson raised possible exceptions in discussion, but the merged text does not spell out a process for granting them.

It is also not clear how the project will determine whether a contribution used AI, what evidence would be considered, or how any contributor-rights sanction would be applied. The proposal warns that these consequences may follow if prohibited use is discovered; the source material does not document a specific enforcement case under the new policy.

Amazon

geographic information system software

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Possible Exceptions Remain Open

QEP-428 is merged, and the stated prohibition is now recorded in the project’s enhancement-proposals repository. Contributors should follow the policy’s stated restriction when submitting code to qgis/qgis. The next meaningful developments would be any clarifying guidance from QGIS maintainers, a defined process for exceptions, or evidence of how the rule is applied in practice.

Dawson’s comments indicate that the project could consider limited uses later, but no timetable or formal exception mechanism is specified in the proposal. Until the project publishes further guidance, the scope of those possible openings remains unsettled.

Amazon

QGIS plugins and extensions

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

What has QGIS prohibited?

The merged proposal bars LLM-, AI- or agent-assisted code contributions to the qgis/qgis repository.

When did the proposal take effect in the project record?

QEP-428 was merged into the QGIS enhancement-proposals repository on Oct. 7, 2026, after its discussion period and vote.

What happens if AI assistance is discovered?

The proposal says the contribution may be reverted and the contributor may face sanctions affecting their project rights. The source material does not set out a detailed investigation or appeal process.

Can contributors use AI for autocomplete or translation?

Dawson mentioned those as possible cases for future exceptions, but the merged proposal does not establish an exception. The written rule prohibits AI-assisted code contributions.

Source: hn

COLUMBUS DAY / I

Columbus Day / Indigenous Peoples' Day Picks

As an affiliate, we earn on qualifying purchases.

You May Also Like

Unlocking AI Potential With Inkling By Thinking Machines

Thinking Machines has launched Inkling, a 975-billion-parameter multimodal model for text, image, and audio reasoning, available on Hugging Face with high hardware demands.

How AI Uses Particle Geometry Mapping In ‘SINGULARITY’ (FABLE/175)

Exploring how advanced AI employs Particle Geometry Mapping in ‘SINGULARITY’ to create immersive environments, blending art and technology.

What DeepSeek-V4-Flash-High Reveals About AI Performance At A Penny Per Million

New data shows DeepSeek-V4-Flash-High achieves high AI capabilities at a fraction of the cost, with recent post-training improvements boosting performance.

The Compute Concentration Audit: When Sovereign Wealth Funds Notice Three Companies Own the Frontier

Global regulatory audits target the dominance of AWS, Microsoft Azure, and Google Cloud over AI infrastructure, revealing critical dependencies.