Today in AI: Kalshi's Agentic Bets, Microsoft's AI Play, and Infrastructure Surge

Ada and Sam unpack the latest in AI, from Kalshi's booming prediction markets and the rise of the 'agentic web' to Microsoft's ambitious AI lab strategy. They also dive into the infrastructure powering AI's growth and discuss crucial conversations around AI safety and transparency.

In this episode

• Hugging Face CEO on OpenAI Hack: 'We need more transparency'
• Today in AI: Kalshi Bets, Agentic Web, and Microsoft's AI Ambition
• Yann LeCun's AMI Labs Papers Give JEPA a Mathematical Foundation
• Hugging Face CEO on OpenAI's AI Security Breach
• Waymo CEO on AI's Real-World Challenges
• Amazon Steps Up Spokane Wildfire Relief
• AI Inference: 10x Faster Models & Self-Optimization
• Amazon Hits $3T; Apple's Subscription Shift
• CoreWeave and Oracle power AI infrastructure surge as hyperscalers sweep Monday rally
• Kalshi Markets Bet on AI IPOs, Moon Landings

Transcript

Ada: Welcome to Today in AI, I'm Ada.

Sam: And I'm Sam. Today, we're tracking the surge in prediction markets, Microsoft's big AI ambitions, and the infrastructure powering it all.

Ada: Let's kick things off with Kalshi, which is really becoming a fascinating barometer for the AI world and beyond. We're seeing a huge surge in prediction markets on the platform, with people betting on everything from AI IPOs to geopolitical events. It’s a real-time, crowd-sourced look at perceived probabilities.

Sam: Exactly, Ada. And what's particularly interesting is how these markets are evolving. We're not just talking about tech trends anymore. The scope is widening, reflecting a growing public appetite to 'price in' future events, especially those with high uncertainty. It's a very direct form of collective intelligence, for better or worse.

Ada: It certainly is. And speaking of evolving landscapes, we're seeing the emergence of what some are calling the 'agentic web,' with companies like MCP Apps pushing the boundaries of how AI interacts with the internet. This isn't just about chatbots anymore; it's about AI agents performing complex tasks autonomously.

Sam: That's a really significant shift, Ada. The idea of AI agents not just answering queries, but actually executing multi-step processes, interacting with different web services, and even making decisions on behalf of users, is transformative. It moves AI from a reactive tool to a proactive participant in our digital lives. The implications for productivity, but also for security and ethical considerations, are immense.

Ada: Absolutely. It's a double-edged sword that we'll be tracking closely. Now, let's turn our attention to one of the biggest players in the game: Microsoft. Mustafa Suleyman, formerly of DeepMind and Inflection AI, is now at Microsoft, and his strategy to make Microsoft a top-tier AI lab is really gaining traction. He's aiming for Microsoft to be among the top three or four AI research powerhouses globally.

Sam: This is a huge play, Ada. Suleyman's track record speaks for itself, and his focus isn't just on integrating AI into existing products, but on fundamental research and pushing the boundaries of what AI can do. It signals Microsoft's intent to not just be a consumer of cutting-edge AI, but a primary creator. This kind of top-down strategic push, combined with their vast resources, could really reshape the competitive landscape in AI research.

Ada: It's a clear statement of ambition, and one that will undoubtedly attract top talent and accelerate innovation. Moving from ambition to the nuts and bolts of AI, the infrastructure powering this surge is also seeing significant movement. CoreWeave, an AI cloud provider, surged almost twenty percent on Monday after lenders restructured its eight point five billion dollar debt facility and Truist upgraded them to Buy.

Sam: This is a classic 'picks and shovels' story in the AI gold rush, Ada. CoreWeave provides the specialized GPU infrastructure that AI models need to train and operate. Their strong performance, coupled with Oracle's nine point two percent jump on full-year cloud momentum, tells us that the demand for AI compute power is not just sustained, but accelerating. These companies are the backbone of the AI revolution, and their financial health and expansion are critical indicators of the broader market's confidence.

Ada: Indeed. The Nasdaq gaining over two percent on Monday, driven by easing geopolitical tensions, also reflects a broader positive sentiment that's certainly benefiting these infrastructure players. Now, shifting gears slightly, the conversation around AI safety and transparency continues to be crucial. Hugging Face CEO Clem Delangue recently weighed in on a breach involving OpenAI's AI models.

Sam: Delangue's comments highlight a critical ongoing debate, Ada. He emphasized the need for greater AI safety, drawing a distinction between open and closed models, and pushing for more robust regulatory frameworks. This isn't just about preventing hacks; it's about ensuring the responsible development and deployment of increasingly powerful AI systems. The more powerful these models become, the more critical transparency and security measures are for everyone involved.

Ada: It truly is. The incident, and Delangue's reaction, underscore that while we're racing ahead with AI capabilities, we also need to be equally diligent about the guardrails. It's a complex balancing act between innovation and responsibility.

Sam: Absolutely. These conversations are fundamental to building trust and ensuring that AI benefits everyone, not just a select few. It's a challenge we'll continue to face as the technology evolves.

Ada: That's all for today's episode. For full stories and more in-depth analysis, visit startuphub.ai.

Sam: Thanks for tuning in to Today in AI.