Advertisement

Google AI could keep baby food safe

Machine learning can spot defective food that humans might miss.

Google's artificial intelligence technology can help the food industry beyond picking better cucumbers. In one company's case, it could prevent your child from getting sick. Japanese food producer Kewpie Corporation has revealed that it's using Google's TensorFlow to quickly inspect ingredients, including the diced potatoes it uses in baby food. The firm and its partner BrainPad trained the machine learning system to recognize good ingredients by feeding it 18,000 photos, and set it to work looking for visual 'anomalies' that hint at sub-par potatoes. The result was an inspection system with "near-perfect" accuracy, culling more defective ingredients than humans alone -- even with a conveyor belt shuttling potatoes along at high speed.

Kewpie goes out of its way to stress that it's supplementing human workers, not replacing them. The TensorFlow system handles the initial removal, but there are still people waiting to inspect the potatoes afterward. Think of this as a booster that helps Kewpie process food faster without compromising quality.

Kewpie's inspectors looking at diced potatoes

The tech is currently limited to Kewpie's baby food, but the company expects to spread it to other foodstuffs before long. Also, it's not hogging the technology to itself. It wants to offer the AI-assisted inspections to other companies, so the industry at large might see a benefit. There's no guarantee that other companies will keep all their human staff (we could see others using this as a chance to cut jobs), but it could set a higher quality baseline.