Meet India-born Huma Abidi, who spearheads AI for Intel's projects

Intel's AI technologies are being used for numerous applications including faster drug discovery to fight diseases like cancer, finding missing children, and optimising crop yields

As the director of machine learning and deep learning software engineering at Intel Corp, Huma Abidi is responsible for making some of the biggest bets in the field of artificial intelligence (AI) technology and drive its applications at the largest microprocessor maker. At a time when Intel wants to bring more AI technology into all aspects of its business, Abidi sees India playing a key role for many of the company's AI-related bets.

The company’s AI technologies are being used for numerous applications. These include shortening the time to train image analysis models for faster drug discovery and evaluating the purity of water sources using AI techniques along with a digital microscope connected to a computer. The other applications include helping severely disabled people drive their chairs with their facial expressions, and detecting poachers entering wildlife reserves and alerting park rangers in near real-time.

“We have the second largest (design centre) in India. They are leading the Internet of Things (IoT) effort and are involved in the AI part. We will be looking at further collaborations,” says Abidi who lives in Cupertino, a 15-minute drive from Intel’s headquarters in Santa Clara, California, in the US.

Abidi who grew up in Lucknow wanted to be a doctor but was introduced to Computer Science by her mother at a time when the field was just emerging in the country. She later went to the US to study and did her MS in Computer Science from the University of Massachusetts Lowell. After graduation, Abidi joined Intel as a software engineer and has since worked in a variety of engineering, validation, and management roles for more than two decades. 

Abidi is now responsible for deep learning software optimisation for Intel Xeon processors. She led a team that helped bring Xeon AI experiences to market and improved software performance by two orders of magnitude resulting in AI revenue of over $1 billion in 2017.

At a time when China has committed to becoming the world’s leader in AI by 2030, pledging billions of dollars in the endeavour, Abidi is of the view that India needs to scale up its training programmes for masses in the field of AI. One of the best things that has happened to AI, she says, is that it is becoming open source and Intel is collaborating with other tech giants in this area including Google, Amazon and Baidu to make the technology accessible. Intel has made several acquisitions in this space including Israeli driverless car technology firm Mobileye for $15.3 billion, deep learning companies Vertex.AI and Nervana, and Movidius, a small chipmaker that made computer vision processors used in drones and virtual reality devices.

While AI was expected to drive some of the most critical services, there is also a growing concern that it could replicate the prejudices that humans had about gender and race. For instance, Amazon had to shut down an AI recruiting tool that showed bias against women. The e-commerce company also had to face a huge issue when it used machine learning algorithms to decide the locations for its Prime same-day delivery services. It was reported that these ML models eliminated black neighbourhoods in places such as Boston and Chicago in the US, and disallowed these areas the same services as wealthier and white neighbourhoods. Abidi says, one needs to make sure that the data is “inclusive” and “explainable” while training the ML model in order to tackle such problems. For instance, if a loan application of a person was rejected by the ML model, the data needs to explain the reason. 

“All companies and universities are doing a lot of work in this area where they are putting a checklist on anything that's being released, like does it have inclusive data, explainability,” says Abidi. “We are working with Google, Amazon, Microsoft (to tackle the issue). This is a major concern because of its potential to do a lot of harm.”