A municipal utility hired Michelle for $49,000 this year. Here is part of what they were buying, in the language of the contract:
“…to equip staff at all levels with a foundational understanding of AI, emphasizing real-world relevance and psychological safety; to integrate AI tools into performance management processes to save time, reduce bias, and improve developmental outcomes”
Psychological safety and bias reduction. Those are line items in a paid scope of work, not just sentiments in a great mission statement.
I want to spend today digging into that, because something people get wrong about AI consulting is assuming you have to choose between ethics and income.
Michelle’s ethics work is some of her highest-paid work
Michelle is the industrial-organizational psychologist you met yesterday. Look at what she’s actually been signing:
$49,000, municipal utility. Staff training, psychological safety, bias reduction in performance management.
$15,000, community services nonprofit. Upskilling workshops, a technology scan, and “governance advisory services including helping them set up a governance entity and routines.”
$22,500, major international nonprofit. Four weeks, 10 deliverables, and in her words “structured around education first, helping the client’s team understand what ‘AI ready’ means and closing the gaps that matter most before they make architecture decisions.”
$4,000, keynote to the CEO, EVPs and SVPs at the National Association of Broadcasters. Her topic list ends with “data, governance, and responsible experimentation.”
$4,000, international nonprofit, to write their AI usage guidelines before they attempt a formal policy.
None of that is pro bono. None of it is a “values” badge stapled onto a technical project. Governance, guidelines and safety are what the client is paying for.
Here’s how she describes why she came to us in the first place:
“We’re not just out here trying to sell a bunch of things. We are here to try to propel responsible and ethical AI use and services that really help an organization grapple with this seismic shift.”
And the reason she keeps going:
“I can actually talk to a nonprofit food bank and have them see that they can actually implement AI at a reasonable level of money. And it’s just amazing to me to know that I can be part of helping that organization advance its mission.”
Some of Michelle's biggest projects help clients with AI governance and responsibility
“Naturally lead the client from training into discovery, strategy, and governance”
That’s Michelle again, in our members-only Slack server, telling the rest of our students where the work goes. Inside the program, this isn’t a single ethics module you click through on the way to the “good stuff.” It’s threaded across the curriculum, because it’s threaded across the work:
AI Bias, Errors and Perpetuating Social Problems
Understanding AI Data Privacy and Cybersecurity
Creating an AI Governance Policy
Intellectual Property, Copyright and Sourcing of Data
Navigating GDPR and the EU AI Act: A Consultant’s Guide
The Colorado AI Regulation: What it means for consultants and businesses
AI Consulting in Highly Regulated Industries
What Europe Can Teach Us About Responsible AI Consulting
AI Risk as a Consulting Opportunity
Mission Meets AI: The AI Consultant’s Guide to the Nonprofit Sector
Nick, another one of our students, chimed in on where client demand is heading:
“The… thing that we’re seeing is really just this ongoing need for more discussions about AI ethics and AI governance. When is the right time to use AI? When does it make our business better? When do we need to take a step back?… How do we teach teams? How do we train teams?”
Two more students went even further. Lyndsay published a free field guide, AI Governance for the Reproductive Health Movement, for organizations working “without technical staff, without AI expertise, and under significant legal and political pressure.” Her reason:
“I built this because as IWAI practitioners we talk about responsible AI implementation. This is what it looks like when the stakes include patient safety, legal surveillance, and communities that have been failed by systems before.”
The environment question, answered with numbers instead of vibes
Tyler kept getting asked about AI’s environmental footprint by clients, and he was honest with himself about how he’d been handling it:
“I realized I didn’t have a good answer beyond ‘it depends.’”
So he built one: an AI environmental impact calculator that estimates carbon emissions, water usage and energy consumption across the major providers, adjusted by region. He built it for his own conversations, and it turned into a tool he uses in client workshops and discovery calls.
That instinct is what we train. Our lesson on AI spending puts the water and electricity cost of a single prompt into concrete terms you can hand a client, so you can give them realistic context rather than either a shrug or a scare. Our self-hosting lesson covers running models on your own hardware, where the energy footprint is closer to a light bulb than a data center, alongside the privacy and cost reasons to consider it.
You don’t have to be an environmental expert to consult on AI. You do have to be able to answer the question when a client asks it, with real data instead of “it depends.” We make sure you have what you need to do that.
How we connect ethics to engineering
We don’t teach you to build your practice by relying on one AI company. We teach you to build repeatable frameworks and then swap the underlying model as better, cheaper or more trustworthy options appear. It makes everything more flexible, more portable, safer and more valuable for you and your client.
The AI companies are not all equally trustworthy. They differ on data retention, on what they’ll train on, on where your client’s information physically sits, on how they handle government requests, and on how candid they are about any of it. Those differences change, sometimes quickly, and usually without much warning.
If your entire consultancy is welded to one provider, you have no way to respond when that provider does something you disagree with. If you adopt our model-agnostic approach instead, the ability to switch is already baked into everything you create.
That’s why we cover open-weight models you can run yourself, self-hosting on your own hardware, and how to evaluate a provider rather than just adopt one.
Here’s Dan, the student you met last week who quit his 9-to-5 and is now working remotely from the beach with his wife and young kids.
“Rob is not a huge cheerleader [for the AI companies]… He’s a real person trying to help out his students… I couldn’t relate to a multi-billionaire up on stage trying to sell me something.”
We’re not beholden to any particular company. We don’t benefit from you buying Claude vs. Gemini vs. whatever else. We just call it as we see it so you can make the best recommendations to your clients.
The best AI consulting is win-win-win
Paul ran a training session for a small trades business in Sydney. The woman who runs the admin side had told him she felt completely left behind by the AI noise and was drowning in it. So he built the whole session around her actual day instead of the technology:
“The thing that landed hardest wasn’t a slide, it was a lo-fi demo of her own emails getting triaged in front of her… Once you anchor there, AI stops being a threat and starts being a relief. It’s change management more than technology.”
He got paid for that training, she got an afternoon back. The best outcome of an AI consulting project is a win-win-win – something that benefits the client, the client’s team, and the client’s customers. You see the worst outcomes in the news, but IWAI students are quietly out there helping companies adopt AI in a way that’s positive for everyone involved.
Enrollment closes Friday at 11:59 PM Pacific. Everything, including the three levels of support, is here:
PS – Tired of doom and gloom AI news? Join The AI Consultancy Project and engineer your ROI with AI PROOF Process™ – one that’s designed to be compatible with a flexible work schedule, and without sacrificing your humanitarian values.
You, your clients, and the world deserve to know what you’re truly capable of.