Home/ Writing/ Force × direction
● Essay · March 2026 · 9 min read · UNSW guest lecture

Force × direction — what AI actually amplifies in strategy work.

Every strategic problem is a vector. AI gives you almost free force. The work — and the moat — is in the direction. Better to go in the right direction slowly than the wrong direction fast.

Diagram: every strategic problem is a vector. Right direction (yellow arrow) reaches the destination; wrong direction (red arrow) misses it, despite being longer.
From my UNSW guest lecture, April 2026 — AI amplifies force. Apprenticeship taught direction.

Strategy looks like a series of slides, but underneath every slide is a vector. Three components, in this order:

Until 2023, force was the bottleneck. Smart, expensive humans took weeks to gather, model, synthesise, and present. The reason consulting cost what it did was because raw analytical force was scarce.

That bottleneck is gone. With Claude Code or a good agent stack, force is almost free. A first‑draft deck, a benchmarking sweep, a model template — these are now overnight outputs. The economics of force have collapsed.

So what is left?

What's left is the direction. And direction does not get cheaper with AI. If anything, it gets more expensive — because you can now move at full speed in whichever direction you happen to point at.

More force, faster, confidently wrong.

That's the failure mode I see in every team I work with. Junior analysts who used to take a week to grind out an answer can now produce one in an hour. The output looks polished. It moves. It compiles. It even reads well. But if the question underneath was wrong, all that polish does is make the wrong answer easier to ship.

The work the AI can't touch.

The lecture I gave at UNSW broke the work into two layers:

Layer 1 — force. Research synthesis, benchmarking, model templating, first‑draft decks, data gathering. AI replaces this. Not augments — replaces.

Layer 2 — direction. Knowing which question to ask. Reading a room. Telling a CEO something they don't want to hear. Earning trust from a scared client. Making a judgement call when the data is genuinely ambiguous. AI does not touch this layer. Not because the models aren't smart — they are — but because the work is fundamentally about being a person another person trusts to make a call.

Two-layer breakdown: Layer 01 — Force (what AI replaces): research synthesis, benchmarking, model templating, first-draft decks, data gathering. Layer 02 — Direction (what AI can't touch): knowing which question to ask, reading a room, telling a CEO something they don't want to hear, earning trust from a scared client, making judgement calls when data is ambiguous.
The two layers. "They can do 80%. They need the last 20%. That last 20% is direction, not force." — Dr. Trung Nghia Ton, Foundry Health

Where this changes how I work.

If force is free, my entire week reorganises. Mondays used to be about briefing the team and watching the gears engage. Now Mondays are for direction work — sitting with the problem, walking the office, talking to the operator who actually runs the channel. The thinking I used to compress into the gaps between meetings now is the work.

The deliverables come almost as a by‑product. By Wednesday I can produce a deck, a model, a brief — but only because by Tuesday I knew exactly what question I was answering, and I'd already pressure‑tested it with the people who'd live with the consequences.

The teams that flounder are the ones where AI fluency arrived before direction discipline. They've automated the wrong half of the job and called it productivity.

If you take one thing from this Before you fire up the AI, ask: am I sure I'm pointing at the right destination? A wrong vector at full force just gets you to the wrong answer faster — and now with better typography.
— Lucas, March 2026 · drawn from the UNSW BIS guest lecture, April 2026 edition