Quality assurance poses a challenge in game development. Testing games bears with itself a labour-intensive and repetitive manual process. QA teams are bogged down in big virtual worlds and complex game systems, tackling a challenge, which is very difficult to complete with the tools they have available. The process is slow, inefficient and expensive. The results are crunches, mistakes, bug-ridden game releases, and loss of game studio reputation.
AI automation of QA testing is not a new notion. Mostly, however, this boils down to Machine learning assisted solutions, that are very hard to set up, learn slow, require huge data sets to learn, and cannot adapt to everyday changes.
We are using a completely different approach.
Machine Learning | Symbolic AI | |
Learning curve | Slow | Fast |
Amount of data needed for learning | Huge | Any |
Human Editing AI's knowledge | Not possible | Possible |
Adaption to changes | Retrain whole model | Continously learning |
Integration | Complex | Simple |
Using the power of Symbolic AI and learning from small data sets, Filuta AI is utilizing deep space exploration tech to create testing scenarios thoroughly and efficiently. The resulting automatable plans are a QA dream for functional tests and complete and open-ended playthroughs, freeing your development costs and time to developing your game.
Our solution makes a significant difference in how your studio handles QA, and we're eager to explore this opportunity with you. To get started, simply schedule a demo presentation with us.
We were supported by the system project Technological Incubation
The AIPlan4EU project is funded by the European Commission – H2020 research and innovation programme under grant agreement No 101016442