Intel Unveils Agentic AI Architectures for Scale

Intel has unveiled its architectures for Agentic AI, introducing three new technologies: Diamond Rapids, Crescent Island, and Wildcat Lake. These aim to scale AI capabilities across computing spectrums, from data centers to edge devices.

Meanwhile, Microsoft's plans for a data center in Kenya have stalled due to concerns over power capacity and water usage. The proposed data center was to run on geothermal energy and incorporate water conservation technology.

Deepfakes are increasingly targeting teachers, with AI-generated videos and images being manipulated to falsely depict them. This has raised concerns about accountability and support for educators.

UC Berkeley has introduced an AI-assisted project in its introductory computer science course, using an AI coding agent to help students learn programming concepts.

OpenAI is developing AI agents that can perform tasks on behalf of users, accessing and controlling various digital services. This has raised concerns about control and security.

Apple has revealed its AI server hardware, featuring a custom 2U rack-mountable chassis with multiple columns of hardware, designed to maximize airflow and efficiency.

The increasing use of AI is making physical workplaces more important, with companies adapting offices to prioritize human connection, mentorship, and shared experiences.

The World Health Organization (WHO) has highlighted Clalit Health Services' OPTICA model as a global benchmark for AI safety in healthcare, providing a structured approach to evaluating AI systems.

The Authors Guild has expressed concerns about AI companies using published works to train models without permission, arguing this could harm the book ecosystem.

Key Takeaways

Intel unveils Agentic AI architectures: Diamond Rapids, Crescent Island, and Wildcat Lake.
Microsoft's Kenya data center plans stalled due to power and water concerns.
Teachers targeted by deepfakes, raising concerns about accountability.
UC Berkeley adds AI assistance to introductory CS course.
OpenAI developing AI agents that can perform tasks on behalf of users.
Apple reveals custom AI server hardware for efficiency.
AI use makes physical workplaces more important for human connection.
WHO highlights Clalit AI safety model for healthcare.
Authors Guild raises concerns about AI training data usage.
Security concerns around AI agents accessing digital services.

Intel unveils architectures for Agentic AI

Intel has outlined its architectures for Agentic AI, showcasing three new technologies: Diamond Rapids, Crescent Island, and Wildcat Lake. These technologies aim to scale Agentic AI capabilities across the computing spectrum, from high-performance data centers to intelligent edge devices. The architectures are underpinned by Intel Foundry technologies, including the Intel 18A process family and Foveros Direct 3D packaging.

Intel Agentic AI architectures unveiled

Intel has announced its strategic push into Agentic AI with the introduction of three new architectures: Diamond Rapids, Crescent Island, and Wildcat Lake. These advancements aim to balance power, cost, and scalability in AI deployments. The architectures are designed to work together to provide a differentiated, scalable foundation for next-generation AI infrastructure.

Kenya's AI ambitions and water needs

Plans for a Microsoft-G42 data centre in Kenya have stalled due to concerns about available power capacity. The proposed data centre was to run on geothermal energy and incorporate water conservation technology. However, residents in the area are already experiencing water shortages, raising concerns about the impact of the data centre on local water resources.

Teachers targeted by deepfakes

Teachers are being targeted by deepfakes, which are AI-generated videos or images that can be manipulated to make people look like they're doing things they never did. The deepfake epidemic in schools is affecting not only students but also teachers, who are struggling to find accountability and support.

UC Berkeley adds AI assistance to introductory CS course

UC Berkeley has introduced an AI-assisted project in its introductory computer science course. The project aims to help students learn programming concepts with the assistance of an AI coding agent.

OpenAI's AI agent ambitions

OpenAI is developing AI agents that can perform tasks on behalf of users. The agents will be able to access and control various digital services, raising concerns about control and security.

Apple's AI server hardware

Apple has revealed the interior of its AI servers, showcasing a custom 2U rack-mountable chassis with multiple columns of hardware. The servers are designed to maximize airflow and efficiency.

AI makes physical workplaces more important

The increasing use of AI is making physical workplaces more important, not less. Companies are adapting their offices to prioritize human connection, mentorship, and shared experiences.

AI agent security concerns

AI agents can pose security risks if not properly configured. The agents can access and control various digital services, making it essential to implement robust security measures.

WHO highlights Clalit AI safety model

The World Health Organization (WHO) has highlighted Clalit Health Services' OPTICA model as a global benchmark for AI safety in healthcare. The model provides a structured approach to evaluating AI systems before implementation.

AI training data concerns

The Authors Guild has expressed concerns about AI companies using published works to train their models without permission. The organization argues that this could destroy the ecosystem for books.