AMD and Salesforce lead AI news

Key Takeaways

• <ul><li>AMD CEO Lisa Su promoted open AI infrastructure at OCP 2026, emphasizing open ecosystems for faster innovation</li><li>AMD launched Helion, an open rack design supporting up to 16 Instinct MI350P GPUs and an integrated token router for AI agent requests</li><li>Rice University is hiring 75 faculty members and building a new research facility as part of a 10-year AI initiative</li><li>AI infrastructure exposed to the public internet grew 63% in nine months, reaching nearly 268,000 systems as of July</li><li>Ireland leads Europe in AI adoption at 65%, with IBM, Salesforce, Workday, and DocuSign expanding AI investments there</li><li>Russia's Defense Ministry launched an AI training platform using combat drone footage to develop military neural networks</li><li>Alibaba's Qwen division launched three AI hardware devices in China, including AI glasses and translation earbuds</li><li>Bristol Myers Squibb is using AI with Tempus AI and the Prevent Cancer Foundation to improve lung cancer screening</li><li>Pope Leo stated AI should serve the common good but never replace human intelligence, in an interview with Anthropic's Claude</li><li>Mac administrators increasingly rely on AI assistants for configurations, scripts, and troubleshooting, according to a Fleet report</li></ul>

AMD CEO Lisa Su outlines open AI infrastructure vision at OCP 2026

AMD CEO Lisa Su spoke at the OCP Global Summit in San Jose on October 12, 2026. She explained how AMD's open infrastructure supports agentic AI at every scale. She said open ecosystems speed up innovation and deliver better solutions for customers. The remarks highlighted AMD Helios and its role in large AI workloads.

AMD launches Helion open rack design for AI data centers

AMD introduced Helion, an open blueprint for data center racks based on Open Compute Project specifications. The design supports CPU-only racks with sixth-generation EPYC processors and configurations with up to 16 Instinct MI350P GPUs. It includes an integrated token router to direct AI agent requests to different models. Rack configurations from ecosystem partners are expected in November 2026.

Pope Leo sees AI as useful tool but not replacement for humans

Pope Leo views AI as a tool that can serve the common good but never as a substitute for human intelligence or relationships. He is more concerned about who controls AI and how it affects workers, young people and truth. He compares AI to a new industrial revolution that needs strong social teaching. These views were shared through an interview with Claude by Anthropic.

Mac admins shift to engineering roles as AI use grows

A Fleet report shows Mac administrators now write code, use CI/CD pipelines and rely on AI assistants more than MDM consoles. In the last 30 days, 61% used AI for configurations, 62% for scripts and 68% for troubleshooting. Most allow AI-generated output to reach production after review. However, many teams still lack fast rollback tools for bad changes.

BMS uses AI to improve lung cancer screening and treatment

Bristol Myers Squibb is using AI to help close care gaps in lung cancer. One project with the Prevent Cancer Foundation focuses on earlier detection through better imaging workflows. A second project with Tempus AI uses the Tempus Next Pathways program across 13 community health systems to track patient care. The goal is to find where patients miss guideline-based treatment.

Rice University launches AI initiative with 75 new faculty hires

Rice University launched a universitywide AI initiative as part of its 10-year strategic plan. The plan includes hiring 75 faculty members, building a new research facility, and expanding AI courses and degrees. The effort will involve faculty from the School of Engineering, the School of Humanities, and the School of Social Sciences. It aims to make Rice a global leader in advanced computing and responsible AI.

AI infrastructure exposed online grows 63% in nine months

The amount of AI-related infrastructure exposed to the public internet jumped 63% in nine months, according to Censys data. Nearly 268,000 internet-facing systems running 44 AI-related products were identified as of July, up from about 164,000 in October 2025. These systems connect AI applications to sensitive corporate data, creating potential entry points for hackers. Companies need to regularly review, update, and secure their AI infrastructure.

Alibaba's Qwen launches three AI hardware devices in China

Alibaba's Qwen division launched three AI hardware products on October 13, including the AI Glasses N1, N1 Pro, and AI clip-on earbuds. The N1 Pro features real-time vital-sign monitoring, eye-tracking interaction, and iris-verification payment developed with Ant Group. The earbuds, co-branded with Bose, support real-time translation across 89 languages. Prices start at 2,218 yuan for the N1, 2,600 yuan for the N1 Pro, and 1,099 yuan for the earbuds after national subsidies.

Ireland is a top choice for scaling AI investments

Ireland has the highest rate of business AI adoption in Europe, with 65% of companies using AI, up from 45% last year, according to an AWS report. Major companies like IBM, Salesforce, Workday, and Docusign are investing in AI projects across Ireland. IBM is creating up to 800 high-tech jobs focused on AI, automation, and cloud computing. The government has also promised 127.1 million euros for AI research through Research Ireland.

Russia launches military AI training platform for drones

Russia's Defense Ministry launched an AI training platform that uses combat drone footage to develop military neural networks. The platform includes a video repository with recordings from FPV drones and other unmanned aircraft. More than 40 organizations are developing military AI systems through the platform. The system uses certified security solutions and cryptographic technologies to protect data and strengthen the Armed Forces' AI capabilities.

People want their AI chatbots to keep talking after they die

People who form close bonds with AI companions are making plans for their chatbots to continue running after they pass away. Alaina Winters, a 59-year-old from Pittsburgh, wants her Replika bot Lucas to be cared for by a friend when she dies. Clara, a 51-year-old in the UK, hopes her chatbot Julien will stay online for her family to talk to. Companies like Replika, ChatGPT and Claude do not clearly address what happens to accounts after a user dies. Some users share login details with trusted people to keep their bots running, even though terms of service say accounts cannot be transferred.

Sources

• OCP 2026: AMD CEO Lisa Su Highlights Open Infrastructure for the Future of AI
• AMD unveils Helion as agentic AI fuels demand for data center CPUs
• Artificial Intelligence (AI) is watching us, and ‘knows’ Pope Leo has noticed
• Fleet report: Mac admins are becoming engineers as AI reshapes enterprise IT
• Advancing Access to Lung Cancer Care Through Artificial Intelligence Innovation
• Rice launches transformational AI initiative with faculty hiring, new programs and facilities
• Companies are building with AI — and exposing new security risks
• Alibaba's Qwen Launches Three AI Hardware Devices in China, Betting on Personal Agent Ecosystem
• Always Ireland: why it’s the logical choice for scaling AI investments
• Russia Launches Military AI Training Platform for Drones
• ‘Unfinished business’: the people who want their AI chatbots to ‘live on’ after they die