Intel announced that it will support developers with the tools they need to more easily utilize AI and implement AI anywhere, so that they can fully access the opportunities that AI offers.

Intel announced on the 21st that on the second day of Intel Innovation 2023, Intel's annual developer event, Intel Chief Technology Officer (CTO) Greg Lavender gave a detailed introduction to how Intel's developer-centric, open ecosystem philosophy helps everyone access the opportunities provided by artificial intelligence.

Developers seeking to leverage AI face challenges that hinder the widespread deployment of solutions across PCs and edge computing, data centers, and the cloud. To overcome these obstacles, Intel is embracing a software-defined, silicon-accelerated approach built on openness, choice, trust, and security.

Intel provides developers with the tools they need to more easily leverage AI and deploy it anywhere, simplifying the development of secure AI applications and reducing the investment required to maintain and scale those solutions.

“The developer community will serve as a catalyst to help industries leverage AI to meet the diverse needs they face today and in the future,” said Lavender CTO. “AI should be responsibly accessible to everyone. “If AI remains the domain of only a small number of experts, the social value it can provide will be very limited,” he said.

In his keynote address on the second day of Intel Innovation, CTO Lavender unveiled Intel solutions for end-to-end security, including Intel® Transparent Supply Chain to verify the trustworthiness of hardware and Confidential Computing to protect sensitive data in use.

The service, codenamed Project Amber, is the first in a new portfolio of security software and services called Intel® Trust Authority, which provides comprehensive, independent assessment of trusted execution environment (TEE) integrity and policy compliance.

Intel Confidential Computing is available wherever it is deployed: multi-cloud, hybrid cloud, on-premises, and at the edge.

As the Intel Trust Authority software and services portfolio continues to grow, the Intel® Trust Platform serves as the foundation for a secure distributed ledger.

Intel is also working to create an open ecosystem that can be easily built on multiple architectures.

This includes joining the Linux Foundation's Unified Acceleration Foundation (UXL). UXL, an industry-wide group, is dedicated to providing an open accelerator programming model that simplifies application development for cross-platform deployment. UXL is an evolution of the oneAPI initiative. Intel's oneAPI programming model allows code to be written once and deployed across multiple computing architectures, including CPUs, GPUs, FPGAs, and accelerators. Intel will contribute the oneAPI specification to the UXL Foundation to facilitate cross-platform development across architectures.

Intel is also collaborating with leading software companies Red Hat, Canonical, and SUSE to provide Intel-optimized upstream distributions of their enterprise software releases, delivering performance optimized for the latest Intel architecture.

In this announcement, Lavender, along with Gunnar Hellekson, vice president and general manager of Red Hat Enterprise Linux, announced an expanded collaboration where Intel will provide upstream support for the Red Hat Enterprise Linux (RHEL) ecosystem using CentOS Stream.

Intel also continues to contribute to AI and machine learning tools and frameworks, including PyTorch and TensorFlow.

To enable developers to scale performance more quickly and easily, Intel Granulate adds autopilot for Kubernetes pod resource authorization. This capacity optimization tool provides Kubernetes users with automatic, continuous capacity management recommendations. This reduces the investment required to meet cost-performance metrics for container environments.

Additionally, Intel Granulate adds autonomous orchestration capabilities for Databricks workloads, enabling an average of 30% cost savings and 23% faster processing times without any code changes.

Addressing the need to protect AI models, data, and platforms from manipulation, tampering, and theft, Intel announced that it is developing an application-specific integrated circuit (ASIC) accelerator to reduce the million-fold performance overhead associated with software-only FHE approaches.

We will also be releasing a beta version of a cryptographic computing software toolkit that will enable researchers, developers, and the community to learn and experiment with FHE coding. The toolkit, scheduled for general availability later this year as part of the Intel Developer Cloud, includes a set of interoperable interfaces for FHE software development, translation tools, and sample simulators for hardware accelerators.