News
Storage costs are hiking for the first time in years, fueled by AI aspirations. Komrpise COO and Co-founder Krishna Subramanian explains how rethinking data tiering strategies can help keep costs under control – not to mention increasing AI accuracy.
Unpacking how recent progress in scaling active inference is already demonstrating real improvements for distributed control systems in energy management, quantum computing, and autonomous fleet management. Early days, but it could impact the future of more capable, safer autonomous and agentic systems. https://diginomica.com/how-aix-might-be-ushering-new-ai-control-paradigm-interesting-agentic-safety-inplications
News
Keen AI founder and CEO Amjad Karim explains how they are building a foundation model to improve electrical grid management. This has implications for reducing the costs of safely running the power grid and lessons for managing other kinds of infrastructure.
Featured
The TechTarget definitional reference piece on generative AI, written for SearchEnterpriseAI. Became the #1 Google result for "what is generative AI" during the technology's breakout year, and has remained one of the most-read explainers in the field. Read at TechTarget: https://www.techtarget.com/searchenterpriseai/definition/
Featured
A Diginomica essay arguing that AI can be usefully understood through Stephen Gilligan's distinction between generative trance and degenerative trance. The piece traces three generations of AI deployment through that lens, draws a cautionary parallel to the rise and fall of NLP as a popular movement, and ends
Featured
A long-form Diginomica interview with Karl Friston, the neuroscientist whose free energy principle and active inference framework offer a different theory of intelligence than the one driving most current AI investment. Friston rarely sits for journalism interviews; this piece captures him on the record about why scaling compute and energy
Featured
This is peer-reviewed work, co-authored with researchers at TII's Secure Systems Research Centre. The paper surveys how generative AI changes the security and resilience picture for autonomous systems: threat modeling, anomaly detection, adversarial robustness, and the architectural implications of letting generative models sit inside safety-critical control loops. Read
A long-form research report for Signal AI on how risk and reputational signals are reshaping executive decision-making. The shift from periodic risk assessment to continuous reputational sensing is real, AI is the engine making it possible, and the implications travel fast across the boardroom, the CISO function, and the communications
Featured
A research paper co-authored with TII's Secure Systems Research Centre, drawing the distinction between autonomic and autonomous systems and arguing that zero-trust principles, originally developed for enterprise network security, can be adapted to govern multi-agent swarm systems where individual agents may be compromised, resource-constrained, or operating with incomplete
A TechTarget feature on autonomous enterprise architectures, written years before the agentic AI wave broke into the mainstream. The piece traces how organizations were beginning to compose systems that made decisions without human-in-the-loop intervention, and what that meant for governance, reliability, and the slow institutional question of who is accountable
News
Over 44 days of dictating, I produced more than 203,000 words and almost no finished work. What I did produce was a new vocabulary: 'speechos', outpourings, and a felt sense of what the next interface with AI might actually be.