In recent years, there has been a growing interest in using artificial intelligence (AI) to improve recycling efficiency. Various startups have emerged with AI-powered solutions for sorting and managing waste at the municipal and commercial levels. However, Finnish startup Binit is taking a unique approach by focusing on tracking household trash.
Borut Grgic, the founder of Binit, describes their upcoming AI gadget as a “sleep tracker for your trash tossing habits.” The device, which is designed to be mounted in the kitchen, utilizes large language models (LLMs) for image recognition of regular household waste objects. By integrating with commercial LLMs like OpenAI’s GPT, Binit is able to accurately identify and track the items that users throw away.
Interestingly, Binit initially attempted to train their own AI model for trash recognition but found that the accuracy was low. They then discovered that incorporating OpenAI’s image recognition capabilities significantly improved the accuracy of their system. Grgic speculates that the high performance of the LLM could be attributed to the large amount of training data it has been exposed to, which includes a wide range of common objects.
Once an item is scanned by the Binit device, the data is uploaded to the cloud where it is analyzed to provide feedback and insights to users. Basic analytics will be available for free, while premium features will be offered through a subscription model. Furthermore, Binit aims to become a valuable data provider on the types of waste being generated, which could be useful for entities such as packaging companies.
While some may question the need for a high-tech gadget to track household waste, Grgic argues that it is all about changing habits. Despite being aware of the need to reduce waste, people often struggle to act on that knowledge. Binit adopts a transparency and gamification approach to encourage users to transform their ingrained habits. During tests in the US, Binit observed a 40% reduction in mixed bin waste as users engaged with the product’s features.
The Binit app not only provides analytics but also offers information and suggestions to help users reduce their waste. For example, when a user scans a piece of packaging, the app generates a card that not only identifies the item but also suggests alternatives to reduce plastic intake based on the user’s location. Binit also sees potential for partnerships with food waste reduction influencers to further promote sustainable practices.
One might question whether Binit’s solution could simply be implemented as a smartphone app. Grgic acknowledges that some households may prefer using a smartphone app, but others may find value in having a dedicated hands-free trash scanner. For this reason, Binit plans to offer both options.
Currently, Binit has been conducting pilot tests in several cities across the US and Europe. They are striving towards a commercial launch in the fall, with a target price of $199 for the AI hardware. Grgic believes this price point aligns with the market for smart home devices.
In conclusion, Binit’s innovative approach to tracking household trash using AI technology has the potential to revolutionize waste management at the individual level. By providing insights, feedback, and personalized recommendations, Binit aims to encourage users to reduce waste and adopt more sustainable habits. As sustainability becomes increasingly important, solutions like Binit’s could play a crucial role in creating a more environmentally conscious society.
Source link
