Platforms, Data Rights, and the Rise of Agentic AI: Who’s in Control?

The Accelerating Pace of AI

• 💡 AI's advent has dramatically sped up technology life cycles, with server life spans shrinking from 7-8 years to under 2 years.
• ⚡ This rapid obsolescence creates significant implications for costs, raw material depletion (e.g., cobalt), and energy demand for data centers.
• 🌱 AI agents offer a solution by enabling predictive analytics for device failures and managing the circular economy through urban mining and recycling.

Data Rights and Governance in AI Consulting

• ✅ Transparency and trust are paramount when using client data for AI-powered consulting, requiring the rewriting of data usage rules.
• 🧠 A significant technical understanding gap exists, making it challenging to explain the difference between anonymized "learnings" and sensitive "artifacts" to executives.
• 🔑 Data ownership remains with the originator, while responsibility for AI actions on that data lies with the customer controlling policy and process.

Edge AI and Distributed Processing

• 🚀 Bringing AI to the data (processing where data lives) is crucial for security, cost, and energy efficiency, especially since most data resides outside public clouds.
• 🌐 Distributed AI processing at edge locations necessitates evolving data rights and governance models to determine accountability for AI inferences.

Agentic AI and Workflow Transformation

• 🧩 Model Context Protocol (MCP) and Agent-to-Agent (A2A) protocols are emerging to facilitate context transfer and interoperability between agents.
• ⚠️ However, these protocols are necessary but not sufficient, as challenges like trust, verification, and data entitlement between agents remain.
• 📈 AI agents are transforming white-collar labor by increasing productivity and enabling a shift to higher-order activities, with humans acting as managers of agents.

Strategic AI Investment and Human Agency

• 🎯 Invest in technologies that solve real problems with clear ROI and defensibility, rather than pursuing AI for its own sake.
• 🔍 Look for companies with unique data access and a vertical focus to create competitive moats against hyperscalers.
• 💡 Ultimately, AI is a tool, and humans retain agency and control over its deployment and use, emphasizing the importance of strategic decision-making.