Game Development Competition For Students Spectacularly Shuts Down Over AI Slop

An Australian competition encouraging young people to make video games is ending after 12 years. Organisers say the rise of generative artificial intelligence has made it increasingly difficult to establish whether entries reflect students' own work or have been produced using AI.

The Australian STEM Videogame Challenge is holding its final competition in 2026. It gave primary and secondary school students a platform to develop games and learn skills. Its closure has disappointed families who valued the opportunity.

Australian STEM Videogame Challenge Ends After 12 Years

Run by the Australian Council for Educational Research (ACER), the challenge invited students to enter individually or in teams of up to four. Participants created games around a chosen theme and submitted supporting material, including a design document and short trailer.

The competition aimed to reward more than a finished game. It encouraged students to plan projects, solve problems, collaborate and understand the creative process behind interactive entertainment.

However, organisers say generative AI has undermined their ability to assess entries fairly. Tools that produce code, artwork and other game elements have made it harder to determine how much work students completed themselves.

Cathy Zemaitis, head of ACER's Student Learning and Progress division, said the decision followed the rapid and complex rise of generative AI in game development. She explained that it had become harder to judge what students made themselves, moving the programme away from its original purpose.

AI-Generated Entries Create Problems for Judges

The concern is not simply that students might use AI as an assistant. It is whether judges can identify work generated largely or entirely by artificial intelligence, rather than created through the skills the competition was designed to develop.

A person who said they had previously judged the competition discussed its closure in a Reddit thread. They described submissions that appeared to have been generated entirely with AI and questioned how judges could fairly assess students' abilities when authorship was unclear.

The issue reflects a wider challenge for educational programmes. AI tools can help students brainstorm, learn coding concepts or experiment with designs, but they can also complete tasks on a participant's behalf. Without reliable ways to verify individual contributions, competitions risk rewarding access to AI rather than students' learning.

Parents and Students Disappointed by Cancellation

One parent on Reddit said their teenage son and his friends had regularly entered the challenge, meeting to discuss ideas and build games together. The competition gave them a shared project outside school and something to work towards.

The parent said their son was visibly upset when he learned the event was ending, describing it as an important part of his identity. They acknowledged that educational programmes must adapt to new technology but felt students were losing an opportunity because of AI-generated entries.

AI Debate Raises Questions About Future Competitions

The closure raises questions about how competitions can embrace technology while assessing students' own abilities.

AI can support prototyping and make game development more accessible. But when a competition is intended to teach coding, design and collaboration, organisers need ways to distinguish assistance from work effectively outsourced.

For 12 years, the challenge offered students a place to experiment, collaborate and share their creations. Its final edition in 2026 will close that chapter, leaving educators to consider how future competitions can protect original student work.