Called Cloud AutoML, the set of services gives businesses automated AI building tools. They can then create machine learning-infused services without a detailed understanding of the underlying tech. Cloud AutoML consumes data supplied by the business to automatically generate a machine learning model. This is then optimised to suit the dataset and begin producing further results.
The initial release of the platform will include one service, Cloud AutoML Vision. This provides a drag-and-drop interface to create new machine learning models using image recognition. Users only need to upload a set of images to start training and optimising an AI model that recognises them. Once the model is confirmed to be working, it can be deployed from AutoML straight to Google Cloud.
Google sees AutoML as a significant development of its AI product portfolio. The company recognised that one of the biggest limiting factors for AI is the lack of experts skilled in the subject. Although demand from businesses is high, there’s evidence that AI deployments are being restricted by the shortfall in developers.
Google’s positioning AutoML to help address this growing problem. Businesses that want to use AI in common scenarios, such as image recognition, could use Google’s self-service automated solution instead of their own in-house team.
READ NEXT: WWF uses blockchain tech to improve transparency in tuna fishing
This could dramatically cut costs and increase the turnaround time when creating new machine learning models. Google said AutoML will move the industry forward by allowing less-skilled developers to start creating complex AI applications.
“Currently only a handful of businesses in the world have access to the talent and budgets needed to fully appreciate the advancements of ML and AI. There’s a very limited number of people that can create advanced machine learning models,” said Google. “To close this gap, and to make AI accessible to every business, we’re introducing Cloud AutoML. We believe Cloud AutoML will make AI experts even more productive, advance new fields in AI and help less-skilled engineers build powerful AI systems they previously only dreamed of.”
Cloud AutoML is the result of a cross-company development effort at Google involving collaboration between the company’s Cloud, Brain and other AI teams. Google said development of AutoML is ongoing and several new products for the platform are being prepared for release.
Businesses interested in the initial computer vision solution can register their interest now. Availability is initially restricted to a select number of users but should be broadened to more Google Cloud customers over time.
