O'Reilly AI Conference: Here’s What You Missed

Equipping businesses to deal with the challenges of Artificial Intelligence.

"The data creation rate doubles every single year and we need to have compute that's going to grow exponentially to keep up with it.”

By Libby Plummer, technology writer

Artificial Intelligence (AI) is accelerating fast and businesses must stay ahead of the curve to avoid missing out. Unlocking the promise of AI was the focus of the recent O'Reilly Artificial Intelligence Conference*, sponsored by Intel. The London-based event comprised four days of keynotes, tutorials, and networking, with the aim of helping the 750 attendees understand how AI will change the business landscape.

As well as helping businesses to unlock the potential of AI, the O'Reilly AI Conference also aimed to equip them to survive the disruption it will bring and find new ways of using AI assets across new industries. The show featured expert speakers form a number of major firms including Microsoft, AWS, HPE, and Google, while Intel also demoed a number of real-world AI applications on its stand, covering everything from predictive analytics in transport to cloud-based speech transcription.

In a keynote talk, Alexis Crowell, senior director of AI Product Marketing at Intel, explained how one of the biggest challenges stopping FSI businesses getting the best from AI is simply not knowing where to start. During an insightful talk, she outlined three key starting points for organizations thinking about implementing AI.

Firstly, Crowell highlighted the importance of being thoughtful, rather than diving straight into overly ambitious and unrealistic projects. "Think through the lower hanging fruits to tackle," said Crowell. "Choose a data problem and stick with it. Choose problems that are small enough to be meaningful and not so big that they’re overwhelming."

Secondly, she focused on security as a key focus. “Security must be at the forefront. Both the data and [AI] model must be secured…make sure the data is correct, needed and can't be tampered with. Thirdly, Crowell explained why transparency should be a fundamental concern. "Transparency in data is all about what's going into the model. It's vital that everyone in the organization is aware of data ethics, asking questions such as 'is it objective and what are we doing with that decision?'", said Crowell. "The biggest barrier to [AI] adoption is employees not understanding and businesses not explaining how they’re using it."

While previous advances in technology have enabled the scaling of output, productivity and access to information, the AI revolution is poised to scale both machine and human knowledge. But in order to gain real business value from AI at increasing scale, organizations must have the right compute technology in place to support it.

"The data creation rate doubles every single year and we need to have compute that's going to grow exponentially to keep up with it," said Crowell. "There are two types of compute that are really needed – DL [deep learning] training and inference."

Recognising the importance of supporting these two types of compute, Intel recently unveiled two new AI accelerators: The Intel® Nervana™ Neural Network Processors (NNP). The NNP-T and NNP-I are built from the ground up with a focus on AI to offer businesses the right intelligence at the right time. The NNP-T is for training deep learning models at scale. The processor is designed to train a network as fast as possible within a given power budget.

Meanwhile, the NNP-I is built for inference and is designed to accelerate deep learning at scale. Both of these processors give businesses the flexibility to cope with ever-increasing amounts of data and the growing scale of AI models, offering near real-time insights and high-volume compute. Not only does this technology offer support for solving today's data problems, it will also be essential in the coming years, where scale will become an increasingly crucial factor.

In order to fully benefit from the AI revolution, FSI businesses must be empowered to use their data efficiently. To do this they must have sufficient compute infrastructure in place. Only then can they successfully deploy enterprise-scale AI with clarity, security, and transparency.

For More Information:

Stay Connected

Stay connected to technologies, trends, and ideas that are shaping the future of the workplace with the Intel IT Center.

By submitting this form, you are confirming you are an adult 18 years or older and you agree to Intel contacting you with marketing-related emails or by telephone. You may unsubscribe at any time. Intel’s web sites and communications are subject to our Privacy Notice and Terms of Use.

By submitting this form, you are confirming you are an adult 18 years or older and you agree to Intel contacting you with marketing-related emails or by telephone. You may unsubscribe at any time. Intel’s web sites and communications are subject to our Privacy Notice and Terms of Use.

Intel Is Enabling the Data Driven Digital Transformation