As Artificial Intelligence (AI) and Machine Learning (ML) get set to take a giant leap in improving day-to-day life, the key is to democratise these new-age tools for all and benefit the communities of developers, users and enterprise customers, a top Google executive said on Wednesday.The concept of AI and ML came into existence long back but with the vast availability of data today, sectors like healthcare, banking and retail are adopting the technologies at a faster pace than before.Google, a pioneer in AI, has been focusing on four key components - computing, algorithms, data and expertise -- to organise all the data and make it accessible."What it entails to democratise AI, we focus on these four core components. "Computing is the backbone of AI technology . Google as a company has always been at the forefront of computing AI," Fei-Fei Li , Chief Scientist of Google Cloud AI and ML, told reporters during a press briefing."We want to make it all accessible to our customers," added Li, also Professor of Computer Science at Stanford University in the US.Earlier this year, Google announced the second-generation Tensor Processing Units (TPUs) (now called the Cloud TPU) at the annual Google I/O event in the US."We announced the Cloud TPU -- the second-generation of our processing unit and our intention is to make it available via Google Cloud," the top executive added.The company offers computing power including graphics processing unit (GPUs), central processing units (CPUs) and tensor processing units (TPUs) to power machine learning.The "Shazam" is one such app that uses GPUs on Google Cloud. The application uses GPUs to match snippets of user audio fingerprints against their catalogue of over 40 million songs.That means when a user "Shazams" a song, the algorithm uses GPUs to search the database until it finds a match for the audio snippet the person has recorded on the phone.Li, however, said that AI still remains among the most complex and new fields. "To make it accessible for businesses and customers where they are required to gain access to the right tools, whether it is a ML library like "TensorFlow" or tapping into pre-trained models via API," Li added.