CTNET

I first noticed this paradox last September — I was listening to a podcast about LLMs and something clicked: I was using AI more, but thinking less independently.

That one observation grew into a 5-part series on AI and human cognition. If you've had a similar moment, Post 1 is a good place to start.

https://www.ctnet.co.uk/the-ai-cognitive-ability-paradox/

#AI #PKM #KnowledgeManagement #ArtificialIntelligence #Writing

When did you last notice AI doing your thinking for you?

I've been exploring this in a new 5-part series on AI and human cognition. The uncomfortable truth: the more we lean on AI, the less we actually think.

What's your experience? Do you tend to question AI outputs, or accept them more often than you'd like to admit?

#AI #PKM #CriticalThinking #KnowledgeManagement #ArtificialIntelligence

"Outsourcing your self-improvement to AI is like getting a robot to exercise for you." — Michael Lim

This is the heart of the problem I explore in my latest post. Cognitive effort builds cognitive muscle. When AI takes over the thinking, we lose exactly the exercise our brains need.

https://www.ctnet.co.uk/the-ai-cognitive-ability-paradox/

#AI #PKM #KnowledgeManagement #CriticalThinking #Productivity

There's a real tension at the heart of the AI age: the better you are at critical thinking, the better you'll use AI. But heavy reliance on AI erodes those very same skills.

Microsoft research confirms it: AI reduces cognitive effort and produces less diverse thinking.

New post: https://www.ctnet.co.uk/the-ai-cognitive-ability-paradox/

#AI #KnowledgeManagement #CriticalThinking #PKM #ArtificialIntelligence

One of my biggest takeaways from "Smarter Than Us": the efficiency trap.

Cut humans out of the loop for speed, and we lose the skills we handed over. Armstrong saw this in 2014. I see it in how people use LLMs in 2026 — taking answers at face value without questioning them.

That's partly why I built an AI literacy framework for my own knowledge system.

https://www.ctnet.co.uk/key-takeaways-of-stuart-armstrong-smarter-than-us/

#PKM #AILiteracy #CognitiveOffloading

Stuart Armstrong argued that creating safe AI requires solving moral philosophy — something humans have been working on for millennia without resolution.

If we can't agree on ethics among ourselves, how do we encode it into something that might become smarter than us?

I'd love to hear your thoughts. Full post here: https://www.ctnet.co.uk/key-takeaways-of-stuart-armstrong-smarter-than-us/

#AIEthics #AISafety #Philosophy

Armstrong warned about an "AI supercommittee" — networked AI instances working together. In 2014, that was speculative.

In 2025, I asked Claude Code to index my notes. It spun up multiple agents, mapped my vault, and created a reusable skill in under 20 minutes. No human committee works that fast.

The future Armstrong described isn't coming. It's here.

https://www.ctnet.co.uk/key-takeaways-of-stuart-armstrong-smarter-than-us/

#AI #AIAgents #FutureOfAI

New post: My key takeaways from Stuart Armstrong's "Smarter Than Us" — a 2014 book on AI risk that reads like it was written yesterday.

I read it in 2025 and merged Armstrong's warnings with what I've experienced using AI day-to-day. From AI agents to the moral philosophy problem, the gap between prediction and reality has narrowed dramatically.

https://www.ctnet.co.uk/key-takeaways-of-stuart-armstrong-smarter-than-us/

#ArtificialIntelligence #AIRisk #BookReview #AISafety

I've documented my AI literacy framework for PKM - three levels addressing strategic delegation, critical refinement, and Socratic interrogation.

Built on cognitive offloading research + 5 years Zettelkasten practice.

What are your red lines? Where do you see AI helping vs. hindering?

https://www.ctnet.co.uk/beyond-fear-building-ai-literacy-for-a-smarter-pkm/

#PKM #Zettelkasten #AILiteracy #Obsidian #DigitalGarden #CognitiveScience

Current experiments: Gemini CLI and Claude (desktop + Code) with direct vault access via MCP.

Early observations:
- Context matters - AI with vault access understands my existing thinking
- Different models have different strengths
- Question quality determines everything

But am I becoming uncomfortable writing without AI? That's a warning sign.

#Obsidian #AI #PKM #Claude #Gemini #MCP

Level 2: Socratic Interrogation - where AI becomes a thinking partner.

The workflow:
1. Develop thinking independently
2. Ask AI to challenge assumptions
3. Ask AI to identify gaps
4. Ask AI for alternative perspectives
5. Synthesise back into your framework

AI input is the start of deeper thinking, never the end.

#CriticalThinking #PKM #AILiteracy #Zettelkasten #SocraticMethod

Concerns about AI and cognitive offloading are grounded in research:
- GPS use reduces spatial memory
- Calculator dependence affects numerical cognition
- Spell-checkers reduce orthographic learning

But intentional use patterns can mitigate these effects.

That's what AI literacy means: understanding trade-offs and making conscious choices.

#CognitiveScience #AIResearch #PKM #DigitalLiteracy

Level 0 of my AI literacy framework: Strategic Delegation

Identify low-value admin tasks and delegate them to AI, freeing cognitive energy for high-value work.

Examples from my practice:
- Tag suggestions
- Formatting checks
- Repurposing opportunities

Time saved goes toward synthesis and connection-making.

#ProductivityTools #PKM #AItools #Obsidian #Zettelkasten

The AI cognitive ability paradox: beginners need cognitive skills to use AI effectively, but AI can undermine those skills during development.

That's why my framework has "red lines":
- Literature notes = human
- Permanent notes = human
- Primary backlinks = human

The friction in these tasks builds cognitive capacity.

#CognitiveScience #AIethics #PKM #Zettelkasten #CriticalThinking

New blog post: Beyond Fear - Building AI Literacy for a Smarter PKM

After 5 years with Zettelkasten and 2+ years experimenting with AI, I've developed a three-level framework for using AI without losing the cognitive skills that make PKM valuable.

It addresses strategic delegation, critical refinement, and Socratic interrogation.

https://www.ctnet.co.uk/beyond-fear-building-ai-literacy-for-a-smarter-pkm/

#PKM #Zettelkasten #AILiteracy #KnowledgeManagement #ArtificialIntelligence #Obsidian

You might be asking yourself,"Why the red lines? Why can't AI write literature notes?"

Because that's where the learning happens.

When you force yourself to write a note in your own words, you're doing the cognitive work that transfers knowledge into long-term memory. Let AI do that step and you've outsourced the learning itself.

The framework exists to help you use AI's power whilst keeping the friction that makes thinking stick.

#AI #Learning #PKM

Practical workflow tip from Monday's AI PKM framework post:

When I have an insightful AI conversation on my tablet, I save it to my vault's ZZZZ_AINotes folder for future reference.

Saturday mornings I review these notes. Most get archived. A few become fleeting notes and enter my normal Zettelkasten processing.

This keeps AI insights quarantined until I've critically evaluated and rewritten them in my own words.

Full framework: https://www.ctnet.co.uk/ai-knowledge-frameworks-level-1-2/

#PKM #Zettelkasten

Yesterday's question that's still sitting with me:

How do you know if you're using AI to *augment* your thinking vs *offload* it?

Warning signs I'm watching for in my quarterly review:
- Rarely modifying AI-generated content
- Can't remember disagreeing with AI output
- Notes starting to sound generic/AI-like
- Feeling uncomfortable without AI assistance

What warning signs would you add?

#AI #CriticalThinking #PKM

I want my AI to push back on me. To challenge my thinking, not just validate it.

That's why "give honest and constructive feedback" is in my personal preferences for every AI model I use regularly.

In today's post, I explain how I use AI as a thinking partner (Level 2) and for critical refinement (Level 1) without letting it do my thinking for me.

The full AI PKM framework: https://www.ctnet.co.uk/ai-knowledge-frameworks-level-1-2/

#AI #PKM #CriticalThinking

New on CTNET: AI PKM Framework Levels 1-2

Level 1 = Critical Refinement
Use AI summaries as starting points, never endpoints. Still engage with source material and write your own notes.

Level 2 = Socratic Interrogation
AI as thinking partner, not answer machine. Challenge assumptions, explore alternatives.

The red lines from Level 0 still apply.

https://www.ctnet.co.uk/ai-knowledge-frameworks-level-1-2/

#AI #PKM #Zettelkasten #PersonalKnowledgeManagement

Hot take: AI summaries are making us worse thinkers.

But they don't have to.

Tomorrow I'm sharing Levels 1-2 of my AI PKM framework—how to use AI for critical refinement and Socratic thinking without cognitive offloading.

The trick? Keeping the right friction in place so you still do the actual thinking.

5pm GMT tomorrow 🧠

#AI #PKM #CriticalThinking #Zettelkasten