Guest: Arun Gupta (LinkedIn)
Company: Intel (Twitter)
Show: Let’s Talk

Although interest in AI has been rekindled with new innovations around generative AI, there are still concerns over the security and ethics of the technology. Intel is one of the major companies working to put the right guardrails in place and helping to develop open frameworks so that AI can be consumed safely. The company is also focusing on lowering the barrier to entry so that more people can benefit from AI.

In this episode of TFiR: Let’s Talk, Arun Gupta, Vice President and General Manager for Open Ecosystem at Intel, talks about some of the key concerns about AI and Intel’s involvement in tackling these concerns. He discusses Intel’s dedication to open source and some of the key projects and initiatives they are working on to help people embrace AI.

Key highlights from this video interview:

• Gupta talks about some of the discussions he saw at KubeCon such as how cloud-native enables generative AI, which he likens to the Linux moment for Kubernetes. He goes on to discuss some of the key discussions coming out of the AI/ML Working Group in CNCF and a strategy meeting they had within the board.
• Securing AI, confidential AI, and how to manage those vulnerabilities are some of the key topics coming out of the Open Source Summit in Tokyo and Open SSF Day. Gupta talks about Intel’s involvement on the boards of open-source projects like PyTorch, the reasons they contribute, and how it helps their customers.
• There are a lot of concerns over the safety of AI and Gupta discusses the challenges of hallucinations of LLMs and how crucial it is to clean up those hallucinations. He explains how this feeds into ethical AI, why having the regulations set out in the Biden Executive Order is important, and how open sourcing these LLMs can help build transparency.
• When we think about open source, it often comes back to the four freedoms but Gupta tells us that the only thing open in Open AI is Open API. He talks about the importance of the AI Alliance to create these much-needed open standards and build transparency and how they are working with the Open Standards Institute.
• While AI has been around for many years, its interest has been reignited with generative AI. Gupta talks about how it is being received by people, some feel threatened while others embrace it. He feels that people should recognize that it can make our jobs easier and that we need the right guardrails to enable us to consume it safely.
• Many people are concerned that generative AI will take away their jobs and Gupta likens it to his experience in the 1980s in India when the Prime Minister of India wanted to bring computers into India but the result was that it automated people’s menial tasks. He feels that the technology is evolving quicker and early adopters will be the winners.
• Gupta discusses Intel’s role in the modern world of AI saying they look at it in four different layers: the foundational silicon hardware layer, the cloud layer where they work to simplify the AI infrastructure, simplifying the AI infrastructure on top of that, and unlocking that AI working with different partners and verticals so that customers can understand the advantage of Intel and leverage it.
• Gupta discusses how Intel is lowering the barrier to entry, saying that they are making it easier for developers to build generative AI applications using PyTorch since they can use the natural data types that are intuitive for developers. He talks about how Intel is working with partners to simplify things for their customers.

This summary was written by Emily Nicholls.