(This was first posted to my LinkedIn profile on Feb 1, 2026)
☕ A slow weekend morning and I was mulling over my work – not so much “am I doing the right thing” but more positional – how does my work fit together with my political beliefs and the changes I want to see in the world. I ended up surfacing an interesting use of genAI – not to get answers, not to reason, not to produce anything, and definitely not to increase efficiency – in fact the opposite – AI as a tool to slow down my thinking.
(OK so maybe a human thought partner would work better, but this was pretty self-indulgent thinking – that’s a big ask of a friend, and an expensive ask of a coach!)
What did I find helpful… I basically went through a structured set of mini-conversations: 🧭 “How do different actors in an ecosystem work on different things to have impact greater than their individual work”, 🧭”Where does my work sit in relation to concepts like ‘reform from within vs. disrupt from the outside'”, 🧭”Who else is working in this space, and how does my work relate”, 🧭”What is my actual position on Big Tech, power, accommodating vs refusing to engage”, 🧭”Does my work match what I say I believe?”, 🧭”What are my criteria for who I will and won’t work with?”, 🧭”How might my ideas travel beyond the immediate outputs of a project”
Some lenses and frameworks that AI threw at me that resonated:
🔹 Complementary roles in social change and movement theory (policy, mobilisation, critique, practice)
🔹 How the age-old ‘Social reform vs revolution’ appears in my context
🔹 Gramsci’s war of position vs war of manoeuvre
🔹 How ideas travel through institutions (legitimacy, norms, indirect paths)
🔹 Audience vs. outputs as a political choice (i.e. is what I am producing the important thing, or who I am producing it for)
🔹 Incentives vs intentions
✅ This left me feeling clearer on my role, my boundaries, who I want to collaborate with and why, and most helpfully clearer on the trade-offs I am prepared to make and those I am not.
❓ What I do next with this I’m still working out, but the idea of explicitly tracing how my work could travel further beyond its immediate outputs (and who I should be collaborating with to make this happen), is definitely appealing.
💬 What about you? I’m enjoying the idea of Slow AI as another use-case… If you’re similarly introspective and this strikes a chord – I have a short conversation starter pack I can share – contact me if you’d like it.
Matt Haikin 