🎮 The Next Input — Issue #170

In partnership with

⚡ The Briefing — 60 sec

• OpenAI ends Microsoft legal peril over its $50B Amazon deal Let’s just be friends again? OpenAI has reportedly won concessions from Microsoft that remove exclusive rights and clear legal risk around the AWS deal, which basically means the hyperscaler chessboard just got rearranged again. (techcrunch.com)

• China blocks Meta’s acquisition of AI outfit Manus Ouch. You gotta feel for the team. Such a large payday… gone. China blocking Meta’s Manus deal is a brutal reminder that in AI, geopolitics can walk straight into the cap table and flip the table. (theregister.com)

• NAB sets up new AI team to guide workforce transformation It sounds good, but if banks are treating AI like they did POCs in the past? Then this will be the first time we see major banks get left behind. AI transformation needs product velocity, not another “innovation theatre” folder. (hcamag.com)

🛠️ The Playbook — The AI Deal Risk Engine

Mission
Help organisations evaluate AI partnerships, vendor dependencies, and internal transformation programs before they get trapped by contracts, politics, or slow-moving theatre.

Difficulty
Intermediate

Build time
3–5 hours

ROI
Cleaner platform bets, stronger negotiating position, and fewer AI initiatives that look strategic but move at bank-speed into irrelevance.

0) Why This Matters

AI strategy is not just technical anymore.

It is commercial, geopolitical, and organisational.

OpenAI loosening Microsoft’s grip so it can work more freely with AWS shows how fast alliances can shift when compute, distribution, and revenue are on the line. China blocking Meta’s Manus acquisition shows that a deal can look done until national interest walks in wearing steel caps. NAB setting up a dedicated AI science team sounds like the right move, but the danger is the familiar enterprise trap: big ambition, careful governance, twelve working groups, and no shipped product.

So the move is not:

• pick a vendor and hope

• run another POC

• let strategy live in committee

• assume big-company AI teams equal execution

The move is:

• map the dependency

• test the commercial risk

• ship small but real workflows

• measure whether the organisation is actually transforming

1) Architecture

2) Workflow

1. List every major AI vendor, platform, and internal initiative currently in play.

2. Map the commercial dependency: cloud, model access, data rights, exclusivity, pricing, and fallback options.

3. Identify which risks are legal, geopolitical, technical, or organisational.

4. Pick one real workflow that can ship in weeks, not quarters.

5. Run a controlled pilot with actual users, not a slide-deck audience.

6. Decide whether to expand, renegotiate, hedge, or kill the initiative based on evidence.

3) Example Prompts

AI Deal Risk Prompt

You are reviewing an AI partnership or vendor deal.

Assess:
- commercial dependency
- cloud or infrastructure lock-in
- exclusivity risks
- data rights
- geopolitical exposure
- fallback options

Return:
1. deal summary
2. top 5 risks
3. questions to ask before signing
4. recommendation: proceed, renegotiate, hedge, or reject

POC-to-Product Prompt

You are converting an AI proof of concept into a real shipped workflow.

For the POC below:
- identify the actual user
- identify the workflow it improves
- identify what must be true for production use
- identify why it might stall
- propose the smallest shippable version

POC:
[insert POC here]

Transformation Theatre Prompt

You are reviewing whether an AI transformation program is real or theatre.

Check:
- whether there are shipped workflows
- whether users are actually using them
- whether governance is enabling or delaying delivery
- whether metrics track outcomes or activity

Return:
1. theatre signals
2. real progress signals
3. one fix

Vendor Hedge Prompt

You are assessing AI vendor concentration risk.

For the stack below:
- identify where we depend too heavily on one provider
- identify what breaks if pricing, access, or terms change
- identify fallback vendors or architectures
- recommend a hedge plan

Stack:
[insert stack here]

4) Guardrails

• Do not confuse a strategic partnership with strategic control.

• Never let a POC count as transformation.

• Keep vendor fallback options visible before contracts get sticky.

• Treat geopolitical exposure as a real risk for AI deals.

• Measure shipped workflows, not steering committee activity.

• If governance slows delivery without improving safety, redesign governance.

5) Pilot Rollout — 3 hours

1. Pick one AI vendor deal, internal AI team, or transformation initiative.

2. Map its dependencies across cloud, model, data, legal, and workflow layers.

3. Identify the top five risks that could block delivery or create lock-in.

4. Select one workflow the initiative can ship quickly with real users.

5. Build a simple scorecard covering risk, adoption, speed, and business value.

6. Review after two weeks and decide whether to expand, hedge, renegotiate, or kill.

6) Metrics

• Number of AI vendors mapped

• Percentage of critical workflows with fallback options

• POC-to-production conversion rate

• Time from idea to shipped workflow

• User adoption rate

• Commercial lock-in score

• Number of initiatives killed before becoming expensive theatre

Pro Tip: The AI teams that win will not be the ones with the best strategy deck. They will be the ones that ship useful workflows before the next committee meeting.

🎯 The Arsenal — Tools & Platforms

• Airtable · map vendors, risks, owners, and fallback options · Airtable

• Google Sheets · fast scorecard for lock-in, risk, adoption, and delivery velocity · Google Sheets

• ChatGPT / Claude · useful for vendor review, workflow mapping, and transformation diagnostics · ChatGPT · Anthropic

• Procurement risk checklists · boring, necessary, and suddenly very relevant when AI deals cross cloud, data, and geopolitics

• POC-to-production trackers · the difference between “we’re doing AI” and “people actually use this now”

Copy-paste prompt block:

You are helping me build an AI Deal Risk Engine.

For the AI partnership, vendor, or internal initiative below:
1. identify the strategic goal
2. map cloud, model, data, legal, and workflow dependencies
3. identify lock-in or exclusivity risks
4. identify geopolitical or regulatory risks
5. identify whether this is likely to ship or become theatre
6. recommend expand, hedge, renegotiate, or kill
7. define the key metrics to track

Initiative:
[insert initiative here]

Return the answer in markdown with sections for:
- Strategic goal
- Dependency map
- Deal risks
- Delivery risks
- Theatre signals
- Recommendation
- Metrics

💡 Free Office Hours

If your AI strategy is getting tangled in vendors, cloud deals, committees, or POCs that refuse to grow up, I run free office hours to help map the risk and turn the thing into a shipped workflow.

Book here: https://calendly.com

Stay in flow state. Dictate everything else.

Context switching kills your focus. Every time you stop coding to type a Slack reply, write a ticket, or draft a PR description, it takes 23 minutes to get back in the zone.

Wispr Flow lets you dictate all of it without leaving your editor. Speak your response, your ticket, your commit message — Flow formats it and you're back to coding. Works system-wide inside Cursor, VS Code, Warp, Slack, Linear, and every app.

4x faster than typing. 89% of messages sent with zero edits. Used by engineering teams at OpenAI, Vercel, and Clay.

Start flowing free

🕹️ Game Over

AI deals are moving fast. Committees are not. Choose your fighter.

— Aaron Automating the boring. Amplifying the brilliant.

Subscribe: link