#digital transformation mindcandy: Recursive language models (RLMs)

The future is future fitness. The advantage isn’t prediction. It’s how fast your team can notice change and respond. Pick one mindcandy prompt below. Discuss it. Decide your moves.  If you want you a repeatable operating rhythm to sense signals early, decide priorities faster, and act with real strategic options, click here.

Today’s prompt: MIT’s new ‘recursive’ framework lets LLMs process 10 million tokens without context rot

By treating the “long prompt” like an external environment the model can inspect, break down, and recursively process—rather than stuffing everything into a single context window. If this works in practice, it changes what’s feasible: agents that can traverse entire codebases, contract libraries, support histories, and audit trails without turning into a summarization blender. Discuss this with your team:

  • Where do we hit “long-context pain” today (sales, ops, product, legal, customer support)?
  • What would we automate if an agent could reliably “read everything” end-to-end?
  • What guardrails would we require: budgets, trace logs, eval tests, and human sign-off?

Other prompts:

My latests book about books about AI

Data, acceleration, and the future of intelligence: Lessons from 25 core books about AI, technology abstraction, and consciousness.

https://www.ronimmink.com/product/a-book-about-books-about-ai/

10 top priorities for CIOs in 2026

Security, data, ERP, innovation, governance, workforce transformation, team, amity, trust, scale, and architecture.

https://www.cio.com/article/4117023/10-top-priorities-for-cios-in-2026.html

6 mistakes companies make when using AI for data analytics

Data analytics has emerged as a leading use case for AI in the business setting. Research shows that 43% of organisations have already implemented AI-powered analytics, with one-third scaling these tools across departments.

https://itwire.com/business-it-news/data/6-mistakes-companies-make-when-using-ai-for-data-analytics.html

How decision flows will redefine analytics in 2026

After years of KPI programmes and dashboard roll-outs, more and more organisations are discovering that insight does not automatically produce action. In 2026, what makes or breaks an organisation’s business analytics strategy will be the effectiveness of its decision flows.

https://www.itweb.co.za/article/how-decision-flows-will-redefine-analytics-in-2026/Gb3BwMWaDW5v2k6V

How Do You Manage AI Agents Before Your Tech Stack Spins Out Of Control

Why agentic ecosystems quickly become chaotic. Businesses need to take practical steps to ensure their agentic ecosystems remain cohesive, secure and stable, as they expand across a growing number of use cases, business processes and teams.

https://www.forbes.com/sites/bernardmarr/2026/01/14/how-do-you-manage-ai-agents-before-your-tech-stack-spins-out-of-control/

AI is finally adding real marketing value but only when humans stay in charge

AI is increasingly integrated into marketing workflows, with its greatest value seen in predictive capabilities that enhance creative development and campaign effectiveness. However, human judgment remains essential to interpret and act on AI insights appropriately.

https://www.kantar.com/north-america/inspiration/ai/ai-is-finally-adding-real-marketing-value

AI slop pushes data governance towards zero-trust models

Organisations are implementing zero-trust models for data governance thanks to the proliferation of poor quality AI-generated data, often known as AI slop

https://www.computerweekly.com/news/366637476/AI-slop-pushes-data-governance-towards-zero-trust-models

What is data optimization?

Data optimization is the process of improving the organization and quality of datasets to ensure efficient data storage, processing and analysis by enterprises and other entities.

https://www.ibm.com/think/topics/data-optimization

A Human You’ll Need in the Loop: The Agentic Engineer

The core enterprise bottleneck is becoming architecture, not adoption. As organizations move from chatbots to agentic systems, the limiting factor is less “getting people to use AI” and more “who can engineer reliable agents.

https://adtmag.com/articles/2026/01/20/the-agentic-engineer.aspx

Reimagining ERP for the agentic AI era

Previous ERP eras fell short of delivering: freedom to innovate outside of vendor roadmaps, capacity for rapid iteration, and true interoperability across all critical functions. Not anymore.

https://www.technologyreview.com/2026/01/20/1129965/reimagining-erp-for-the-agentic-ai-era/

Ron-Immink.jpg

Daily #MindCandy

Subscribe to my (free!) near-daily scenario prompts—designed to spark strategic thinking.


Each edition delivers fresh insights, emerging trends, thought-provoking prompts, and must-read business books to keep your mind bubbling and your strategy sharp.

Scroll to Top
0 Shares
Share
Share
WhatsApp
Email