We get asked all the time: “Do you actually use AI in your own work?” Short answer: every single day in myriad ways. Longer answer: we use AI in the way we advise clients to use it – as an improvement accelerator, not as an autopilot. Here’s a straight‑from‑us look at what that means in practice.
Why We Use AI as an Accelerator, but not as an Autopilot
When we’re thinking divergently, surfacing new ideas or concepts, AI helps us widen the funnel of ideas and get to clarity faster. When we’re thinking convergently, consolidating groups & parts of ideas into synergistic wholes, Large Language Models (LLMs) are great at first‑pass synthesis: turning messy notes into clean summaries, laying out option sets, and highlighting patterns we might want to pressure‑test.
But there’s a line we never cross: we don’t outsource judgment. Tools can draft a dozen possible roadmaps; only humans can weigh trade‑offs, politics, budgets, culture, and timing. Think of AI as the turbocharger that helps cover more ground between meetings, while we keep our hands on the wheel.
How We Keep AI Safe
Speed is useless if trust takes a hit. We run AI inside secure workflows and keep sensitive details out of public tools. When we do use general LLMs, we sanitize inputs—removing names, numbers, contracts, anything confidential. We keep client content in private repositories with access controls and audit trails, and we review every AI‑assisted output before it goes anywhere. “Human-in-the-loop” isn’t just a buzzword; it’s our recommended architecture for AI for most use-cases, (at least through versions 2-3… If you think your AI is ready to go “uber” on full auto-pilot, we’re happy to be the skeptics who can help you smoke-test whether it’s truly ready.)
Data quality – both in terms of accuracy AND completeness – matters just as much as security. Models are only as good as the sources we feed them. So we validate inputs, note conflicts, and document assumptions. If the source is shaky, the answer is at best a starting point, not a conclusion. Our rule of thumb: AI drafts; consultants decide.
Where It Shows Up in Our Day
If you sat with us for a week, you’d see AI pop up in a lot of ordinary moments:
- Interview synthesis. After stakeholder conversations, we use LLMs to produce concise summaries and pull themes we might otherwise spend hours extracting. That lets us get feedback to sponsors faster.
- Assessment scaffolding. During platform and vendor assessments, AI helps us assemble comparison tables, flag mismatches between claims and requirements, and draft risk registers. We still do the vetting; the model just builds the scaffolding.
- Roadmap drafts. First‑pass narratives, dependency lists, and swim lanes come together more quickly. We then rewrite, simplify, and ground each step in the client’s reality.
- Business development. Prospect research tightens up. Proposals start with smart boilerplate but end with tailored language. We spend more time on the unique “why this, why now” and less time retyping the obvious.
- Back office. Meeting agendas, RFP checklists, and status updates, and contact segregations for marketing campaigns get a head start. Humans approve everything; AI just clears the runway.
None of that is flashy. It’s the quiet compounding that happens when minutes saved become hours freed for the human work of alignment, decision‑making, and change leadership.
What Stays Human
The “hard part” of transformation has never been the technical decisions – it’s the people. That remains just as true in the “AI era”.
- Sponsorship and alignment. Securing the right executive backing, framing trade‑offs, and keeping cross‑functional teams on track still takes credibility and timing. A model can’t read a room—or rebuild trust when a project hits turbulence.
- Sequencing and scope. Choosing what to do first, what to postpone, and how to stage value so momentum grows is a judgment call shaped by culture and risk tolerance. Tools can simulate scenarios; leaders choose a path.
- Negotiation and governance. Vendor selection, contract terms, escalation paths, and decision rights are human conversations. AI can draft options; it can’t hold the line.
- Change adoption. Training, incentives, communications, and feedback loops determine whether new processes stick. Empathy beats automation here every time.
Our Value Proposition isn’t Just about Bringing Answers – It’s Mainly About Making Those Ideas Work in the Real World
The value proposition of Innovation Vista Isn’t just (or even mainly) about “bringing answers to the table” – especially in the AI era. It’s about bringing expert leaders to the table who can bring our best ideas to life – the best ideas from both the client and our team.
Consulting firms which sell “a process” and staff it with newly-minted MBAs are being disrupted by AI at every turn these days; looking at you: Deloitte, PwC, Accenture, KPMG, McKinsey, Bain, BCG, even Gartner and Forrester. A quick google on these firms prospects reveals the level of fear they have reached; their old model makes little sense in an age when any executive can prompt a free AI LLM for a process tutorial on any change they’re exploring.
On the other hand, at Innovation Vista we understand that we bring more value with the leadership & communication skills of our consultants – applied through the lens of their industry business knowledge AND deep technical skills – than with any “pre-baked” answers they have in their heads. Collaborative strategy & “answers” are always far more effective and likely to succeed within the organization’s context than anything pre-baked anyway.
The Playbook We Recommend to Clients
We use one governing framework for ourselves and recommend it to clients:
- Start AI where it’s feasible, safe, and clearly ROI‑positive. Good first targets are places with solid data and low downside: knowledge‑base cleanup, service desk summaries, first‑pass analysis, draft communications, vendor comparison matrices.
- Set guardrails early and co-evolve AI governance. Define what data is in‑bounds, how content is anonymized, who can use which tools, and what requires human review. Write it down. Train people.
- Pilot with purpose. Pick use cases with measurable outcomes: faster time‑to‑proposal, reduced rework, shorter cycle times. Decide in advance what “good” looks like.
- Pair the tech with the human work. Every pilot needs sponsor engagement, a change impact check, and a plan for adoption. If you don’t assign owners and design incentives, you’ll get demos—not durable value.
- Let early wins fund the next wave. Use time and dollars you save to spin the flywheel and fund the next project(s). That’s how AI becomes a sustainable improvement engine, not a one‑off experiment.
The goal isn’t “AI everywhere.” It’s “AI where it clearly pays off” – and where it frees your people to raise service levels or reinvent parts of your business.
How We Measure Value
We track three simple metrics to keep ourselves honest:
- Time to clarity. How quickly do we move from raw notes to a credible set of options for a decision? If AI is working, this window shrinks without sacrificing rigor.
- Option quality. Are our scenarios more complete and easier to compare? Do we see fewer blind spots because we explored more angles up front?
- Leader & expert time reallocation. Are our consultants spending more hours on stakeholder work—alignment, decisions, and follow‑through—and fewer on drafting?
Those measures are tangible. You can feel them in better meetings, clearer roadmaps, and projects that start on time because sponsors are aligned before kickoff. If AI isn’t helping there, it’s just noise.
The Future is Bright – Ask Again in 6 Months, We’ll be Using AI in New Ways
We expect AI to keep absorbing more of what used to be “table stakes” consulting tasks: first‑pass research, structured synthesis, and initial drafts. We welcome that. It makes space for the work that actually determines outcomes: judgment, orchestration, and the messy art of change.
Our commitment is simple. We’ll keep using AI responsibly and securely, on quality data, and where the ROI is clear. We’ll stay transparent about how we use it. And we’ll keep putting human leadership at the center, because that’s what turns ideas into results.
If you want the specifics of our AI use policy or you’re mapping your first few AI‑powered wins, we’re happy to share what’s worked for us, and what hasn’t. In the end, the promise of AI isn’t that it writes your strategy for you. It’s that it helps your team reach a good strategy faster – and then gives you back time and attention to deliver it.
