Pope Leo XIV recently addressed the dangers of AI in warfare at Rome's La Sapienza University, warning that advanced weaponry could lead to global annihilation. Speaking on May 14, 2026, he criticized rising military spending and urged better monitoring to ensure human responsibility in conflicts involving Ukraine, Gaza, Lebanon, and Iran.
While global leaders focus on safety, tech giants are racing to solve technical hurdles. IBM engineer Meenakshi Kodati explains that agentic AI systems often fail due to infinite loops and planning errors, suggesting clear tool limits are necessary. Meanwhile, researchers have developed asynchronous processing methods to enable real-time interaction, achieving speedups of 1.6 to 2.2 times for smaller models on edge devices.
Major companies are deploying specific solutions to these challenges. Amazon launched Alexa for Shopping, a personalized assistant combining Rufus's product knowledge with Alexa+ history, available without a Prime membership. NVIDIA introduced the Vera Rubin Platform to handle scaling issues for agentic workloads using extreme co-design across thousands of chips. Microsoft experts Amy Boyd and Nitya Narasimhan highlighted a critical gap in tools for monitoring AI agent reasoning and debugging internal states.
OpenAI supports a US-led global AI governance body that would include China, aiming to set safety standards similar to the International Atomic Energy Agency. This comes as President Donald Trump visits China to discuss trade and chip restrictions. Beyond infrastructure, Joe McKendrick argues that knowledge management is essential for business AI success, noting that poor data leads to underperforming initiatives. In academia, Penn State provost Fotis Sotiropoulos predicts AI will reshape doctoral research by speeding up exploration while emphasizing that human judgment remains irreplaceable. Finally, the AI race is shifting to power supply, with Europe facing grid delays that could push investment to regions with faster compute scaling.
Key Takeaways
['Pope Leo XIV warned at La Sapienza University on May 14, 2026, that AI and advanced weaponry threaten global peace in Ukraine, Gaza, Lebanon, and Iran.', 'OpenAI supports a US-led global AI governance body that would include China to set safety standards.', 'IBM engineer Meenakshi Kodati attributes agentic AI failures to infinite loops and planning errors caused by probabilistic models.', 'Researchers developed asynchronous processing techniques that speed up agentic AI by 1.6 to 2.2 times on edge devices.', 'Amazon launched Alexa for Shopping, a personalized assistant available on the app and website without requiring a Prime membership.', 'NVIDIA introduced the Vera Rubin Platform to solve scaling issues for agentic AI workloads using extreme co-design across thousands of chips.', 'Microsoft experts Amy Boyd and Nitya Narasimhan identified a critical lack of tools for monitoring AI agent reasoning and debugging.', 'Joe McKendrick states that knowledge management is essential for business AI success because poor data leads to underperforming systems.', 'Penn State provost Fotis Sotiropoulos believes AI will accelerate doctoral research but will not replace the need for human judgment.', 'Europe faces a new test in the AI race as grid delays and high costs hinder data center construction despite clean energy resources.']Pope Leo XIV warns AI warfare risks global destruction
Pope Leo XIV spoke at Rome's La Sapienza University on Thursday to warn against the dangers of AI in warfare. He called the trend a spiral of annihilation that threatens peace in Ukraine, Gaza, Lebanon, and Iran. The Pope criticized high military spending and urged better monitoring of AI to keep humans responsible for their choices. He also met with Palestinian students who arrived in Italy via a humanitarian corridor to continue their education.
Pope Leo XIV warns AI and weapons could destroy the world
Pope Leo XIV delivered a speech at La Sapienza University in Rome on May 14, 2026, warning that AI and advanced weaponry lead to global annihilation. This visit marked the first time a pope has spoken at the campus since 2008. The Pope met with Palestinian students who recently arrived in Italy to study and receive medical care. He criticized the increase in military spending and called for education that values life and justice.
OpenAI supports US-led global AI group including China
OpenAI supports the creation of a global AI governance body led by the United States that would include China as a member. Chris Lehane, OpenAI's vice president of global affairs, suggested the group could work like the International Atomic Energy Agency to set safety standards. The idea involves linking the US Commerce Department's AI center with safety institutes worldwide. These comments came as President Donald Trump visited China to discuss trade and AI chip restrictions.
IBM engineer explains why agentic AI systems fail
IBM Advisory AI Engineer Meenakshi Kodati explains why agentic AI systems often fail due to infinite loops, planning errors, and unsafe tool use. These issues happen because the underlying probabilistic models can make mistakes or get stuck repeating actions. Kodati notes that failures often stem from system design flaws rather than just the AI model itself. She suggests defining tool limits clearly and adding validation checks to make these systems more reliable.
New methods unlock real-time interaction for agentic AI
Researchers have developed new techniques to make agentic AI systems work in real-time for complex tasks. The main problem was that AI reasoning and tool use caused long delays in conversations. The solution involves asynchronous processing that separates thinking from waiting for user input. These methods allow smaller AI models on edge devices to achieve speedups of 1.6 to 2.2 times while keeping accuracy high.
Amazon launches personalized Alexa assistant for shopping
Amazon introduced Alexa for Shopping, a new personalized AI assistant available on the Amazon app and website. This tool combines the product knowledge of Rufus with the personal history of Alexa+ to help users shop better. Customers can ask questions, create shopping guides, compare products, and track prices without needing a Prime membership. The assistant remembers user preferences across phones, laptops, and Echo devices to provide tailored recommendations.
Penn State provost discusses AI impact on doctoral education
Fotis Sotiropoulos, Penn State's provost, wrote an op-ed about how AI will change doctoral education and research. He predicts AI will fundamentally shape the research landscape by speeding up the process and expanding what scientists can explore. Sotiropoulos believes AI will not replace PhDs because human judgment, ethics, and curiosity remain essential. He sees this technology as a major opportunity for Penn State and the people of Pennsylvania.
Microsoft experts highlight missing tools for AI agent monitoring
Microsoft experts Amy Boyd and Nitya Narasimhan discussed a critical gap in tools for observing AI agents. As AI agents become more complex, developers struggle to understand their internal reasoning and debug errors effectively. The presentation highlighted the need for better methods to trace decision-making logic and visualize internal states. Bridging this gap is essential for building trustworthy AI systems that can be deployed in critical applications.
NVIDIA Vera Rubin platform solves scaling issues for AI agents
The NVIDIA Vera Rubin Platform is designed to solve scaling problems for agentic AI workloads that require predictable performance. It uses extreme co-design to combine high-speed computing with low-latency networking across thousands of chips. This approach ensures consistent speed for multi-turn requests and large models that need to process vast amounts of data. The platform uses specialized links and compiler scheduling to manage data movement efficiently.
Knowledge management could save AI from poor results
Joe McKendrick argues that knowledge management is essential for the success of artificial intelligence in businesses. Many companies focus on language models but ignore the unstructured data that powers them. He states that AI is only as good as the knowledge it is built upon, and poor data leads to underperforming AI initiatives. Integrating knowledge management into AI strategy is necessary to capture and analyze information effectively.
AI race shifts to power supply as Europe faces new test
The competition for AI leadership is now driven by access to electricity and computing power rather than just data. Europe has clean energy resources but struggles to build data centers fast enough due to grid delays and high costs. Without reform, AI investment may shift to regions where power and compute scale more quickly. Experts suggest aligning clean energy supply with data center demand to secure Europe's digital future.
Sources
- Pope decries the rise of AI-directed warfare, saying it leads to a spiral of annihilation
- Pope Leo XIV warns of AI and weaponry leading to global annihilation
- OpenAI backs creation of global AI governance body led by the U.S. that would include China as a member
- Agentic AI Fails: Loops, Planning & Unsafe Tool Use
- Real-Time Agentic AI Unlocked
- Meet Alexa for Shopping, your personalized, agentic AI assistant on Amazon
- Leadership insights: Provost Sotiropoulos on AI, doctoral education and the land-grand mission
- Mind the Gap in Agent Observability
- How the NVIDIA Vera Rubin Platform is Solving Agentic AI’s Scale-Up Problem
- Knowledge Management, The Tech World’s Step Child, May Be AI’s Salvation
- The AI race is shifting to power — and Europe faces a new test
Comments
Please log in to post a comment.