Global AI Weekly
Issue number: 86 | Tuesday, February 25, 2025
Highlights
}
Nvidia's earnings to be a test of AI chip demand as DeepSeek sows spending doubts
Demand for Nvidia's pricey artificial intelligence chips will be in focus when the company reports results on Wednesday as investors doubt the hefty spending on the technology after low-cost AI models from China's DeepSeek rattled the industry.
reuters.com
}
Grok 3 appears to have briefly censored unflattering mentions of Trump and Musk
When billionaire Elon Musk introduced Grok 3, his AI company xAI’s latest flagship model, in a live stream last week, he described it as a “maximally truth-seeking AI.” Yet it appears that Grok 3 was briefly censoring unflattering facts about President Donald Trump — and Musk himself.
techcrunch.comResearch
}
Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?
The leading AI companies are increasingly focused on building generalist AI agents -- systems that can autonomously plan, act, and pursue goals across almost all tasks that humans can perform. Despite how useful these systems might be, unchecked AI agency poses significant risks to public safety and security, ranging from misuse by malicious actors to a potentially irreversible loss of human control.
arxiv.org
}
Understanding Emergent Capabilities in LLMs: Lessons from Biological Systems
Large language models have emergent capabilities that come unexpectedly at scale, but we need a theoretical framework to explain why and how they emerge. This paper proves that language models are actually non-ergodic systems while providing a mathematical framework based on Stuart Kauffman's theory of the adjacent possible (TAP) to explain capability emergence. Their resource-constrained TAP equation demonstrates how architectural, training, and contextual constraints interact to shape model capabilities through phase transitions in semantic space.
towardsdatascience.comVideo
}
Build your code-first app with Azure AI Agent Service
Tune in on 25th February at 9AM PST (6PM CET) to join the upcoming Reactor live-streaming on building your code-first app with Azure AI Agent Service. Artificial intelligence agents are everywhere—from chatbots and virtual assistants to autonomous systems making decisions on behalf of users. But what exactly is an "AI agent," and how does it differ from other AI models and systems? In this talk, we’ll demystify the concept of AI agents, exploring the core components that define them, their unique capabilities, and how they interact with users and environments in real-time.
techcommunity.microsoft.com
}
Silicon Minds, Human Hearts - Rob Ferguson
In this interview, we sit down with Rob Ferguson, the head of AI for Microsoft for startups, to uncover the fascinating intersection of AI and human innovation in the startup world. From his experience as a CTO of multiple startups to witnessing the evolution of machine learning over the past two decades, Rob shares insights on the unprecedented changes in AI and its impact on our lives. Join us to explore how AI is shaping the future of startups and the potential it holds for groundbreaking advancements.
youtube.comArticles
}
We gave an AI a Rorschach test. What it saw in the inkblots offers a window into the human mind
Rorschach tests play with the human imagination and our mind's ability to impart meaning onto the world around us – but what does AI see in them?
bbc.co.uk
}
‘Engine of inequality’: delegates discuss AI’s global impact at Paris summit
Emmanuel Macron’s tech envoy warns attenders current trajectory of artificial intelligence is unsustainable The impact of artificial intelligence on the environment and inequality have featured in the opening exchanges of a global summit in Paris attended by political leaders, tech executives and experts. Emmanuel Macron’s AI envoy, Anne Bouverot, opened the two-day gathering at the Grand Palais in the heart of the French capital with a speech referring to the environmental impact of AI, which requires vast amounts of energy and resource to develop and operate.
theguardian.comUpcoming Events
}
Global AI Bootcamp 2025
The Global AI Bootcamp is an annual event that occurs worldwide, where developers and AI enthusiasts can learn about AI through workshops, sessions, and discussions. Local chapters of the Global AI community organize this event worldwide in March and April.
globalai.communityCode
}
github.com/PennyroyalTea/gibberlink
A demo of two conversational AI agents switching from English speech to sound-level protocol when they realize they are both AI agents.
github.comPodcast
}
Practical AI: Machine Learning, Data Science, LLM
Making artificial intelligence practical, productive & accessible to everyone. Practical AI is a show in which technology professionals, business people, students, enthusiasts, and expert guests engage in lively discussions about Artificial Intelligence and related topics (Machine Learning, Deep Learning, Neural Networks, GANs, MLOps, AIOps, LLMs & more).
open.spotify.com