Take a position before AI takes one for you.

AI is already in your organization — in the tools you pay for, the features that quietly switched on, and the personal accounts your team uses when no one's looking. Most AI policies were written for data security and say nothing about sustainability. This track shows you how to fix that.

You'll learn how to:

• Surface the shadow AI and invisible features already multiplying your footprint

• Design a policy that gets adopted, not ignored — principles-based, with the operational detail that makes it real

• Set data-handling thresholds, human-in-the-loop rules, and disclosure norms

• Integrate environmental criteria into the AI policy your org probably already has

• Map the stakeholders you need — and name the first five people you have to talk to

You'll walk away with: A position statement ready for your ethics, sustainability, or executive committee — complete with a data-handling matrix, a named owner and governance cadence, and the stakeholder map to get it adopted.

Best for: sustainability leads, legal and compliance, ops leaders, and anyone tasked with "figuring out our AI policy."

You can't reduce what you can't measure. Start measuring.

Hyperscalers don't publish the numbers you need. The GHG Protocol wasn't built for AI. The standards are still being written. This track teaches you to produce a defensible estimate anyway — and turn it into action.

You'll learn how to:

• Navigate what the major AI providers do (and don't) disclose — and why non-disclosure is itself a data point

• Set emission boundaries and avoid the Scope 2 and Scope 3 traps where AI hides

• Build a first-pass estimate of your AI hotspots using a proven methodology

• Rank opportunities by impact × influence — because a list of problems you can't control isn't a plan

• Say what's not worth doing, and defend it

You'll walk away with: A hotspot map of your organization's AI footprint, a defensible first-pass estimate with assumptions stated honestly, and a ranked shortlist of 3–5 feasible moves — plus an improvement path for what can't be known yet.

Best for: sustainability and carbon-accounting practitioners, ESG reporting leads, and anyone who's been asked "what's our AI footprint?"

You're not a hyperscaler and you're not a frontier lab — but you sign the contracts. Most organizations' largest AI impact sits inside a supplier's estate, and the only way to get the data is to ask well. Nobody teaches this. We do.

You'll learn how to:

• Decode the AI vendor stack — who has leverage, who has data, and what to ask each of them

• Read vendor answers like a pro: what a good answer looks like, what deflection looks like, and what silence tells you

• Build weighted criteria that serve your environmental goals and your strategic priorities

• Write the data request into the contract, so it arrives every year — not once during the sales cycle

• Use AI sustainably at scale: model right-sizing, routing, carbon-aware design, and on-device vs. cloud trade-offs

You'll walk away with: A complete diligence pack — vendor question set, weighted criteria, data request schedule, and contract clause starters — pilot-scored on one or two real providers, with the named renewal you'll run it on. Or, for builders: a sustainable AI agent or tool designed with green software principles.

Best for: procurement and vendor management, IT and digital leads, and anyone with a software renewal on the calendar.

• Small peer groups. Feedback, accountability, and cross-pollination — every deliverable is tied to your own organization.

• Real templates. Policy skeletons, questionnaires, scorecards, and clause libraries you keep.

• Honest scope. Eight weeks won't produce a board-approved policy or a complete emissions inventory — and we won't pretend otherwise. It will produce the working instrument that gets you there.