Enterprise AI transformation

Fifteen years implementing transformation inside large enterprises. The last three focused on what frontier AI means for organisations, work, and society.

The safety conversation belongs in the room where transformation decisions get made. It rarely is. That is the gap I work in.

Credentials
15
Years enterprise transformation across finance, procurement, legal and HR
5
Enterprise functions where the framework has been validated in live engagements
Frontier
safety lens
Bringing AI safety research into the room where enterprise decisions get made

"Frontier AI safety researchers and enterprise transformation practitioners are two communities not talking to each other. That is the problem I am here to solve."

Recent insights
Implementation
The five most expensive sentences in AI transformation right now
What leaders say that signals the project is already failing, before a single line of code is written.
Read →
Frontier safety
Why the safety conversation belongs in the transformation room
Frontier AI safety research and enterprise transformation practice are two communities not talking to each other.
Read →
Field observations
Agentic AI and the governance question no one is asking
Organisations deploying agentic systems without asking who is accountable when the agent decides differently.
Read →
All insights →
Professionals and teams
Work through a challenge with me
You are implementing agentic AI inside a large function and something is not working. One to one or up to five people. You bring the problem. I bring 15 years of enterprise transformation and a frontier safety lens.
Small businesses
AI audit for owner-managed businesses
Fixed-price AI audit for businesses of 5 to 50 people in Wandsworth and South West London. Practical implementation support after. No jargon, no platform agenda.
Speaking
I speak on frontier AI and enterprise reality

Conferences, panels, podcasts, internal knowledge sessions and professional networks. The safety lens throughout.

AI implementation beyond the hype
What makes enterprise AI transformation fail, and what the research says about why.
Why frontier safety matters for business adoption
What safety researchers know that transformation directors do not and should.
From pilots to operating models
The failure modes between proof of concept and production that no one talks about.
Agentic AI and the future of work
What agentic systems mean for accountability, governance and the human in the loop.
The gap between AI labs, policy and the enterprise floor
Why the conversation in research is not reaching the people who need it most.
About

I work at the intersection of frontier AI safety and enterprise transformation.

Fifteen years implementing large-scale transformation inside enterprise functions. Finance, procurement, legal, HR, controllership. Not advising from the outside. Building from the inside, in the room where the decisions get made.

For the last three years that work has been inseparable from AI. Not the version of AI that gets announced at conferences. The version that lands inside a large organisation and immediately runs into governance gaps, framing failures, platform decisions made too early, and facilitation rooms without enough seniority to challenge what is being proposed.

The frontier AI safety community is producing some of the most important thinking about where this technology goes and what it means for society. The enterprise transformation community is deploying it at scale, today, with real consequences for real organisations. These two communities are not talking to each other. That is the problem I am here to solve.

I built a framework from live engagements — five phases, each with named failure modes. It applies across functions and sectors. I use it to help transformation directors implement agentic AI without the most common and expensive mistakes.

Experience
15 years enterprise transformation
Functions
Finance, procurement, legal, HR, controllership
Focus
Frontier AI safety meets enterprise reality
Based
London, UK
The framework

Agentic AI Implementation Framework for Enterprise Functions

01
Framing
Failure mode: reinvention vs automation mindset
02
Platform selection
Failure mode: CIO not in the room
03
Discovery
Failure mode: no structured process templates
04
Developer selection
Failure mode: wrong experience profile
05
Facilitation
Failure mode: not senior enough to challenge
Want to work through something with me?
Insights

Practitioner thinking on frontier AI and enterprise reality

Frameworks, failure modes, field observations. Written for transformation directors who want the unfiltered version.

All
Implementation
Frontier safety
Field observations
Implementation
The five most expensive sentences in AI transformation right now

What leaders say that signals the project is already failing, before a single line of code is written. These are not edge cases. They appear in almost every enterprise AI engagement, across every function, at every level of seniority.

Read full article →
Also in implementation
What the discovery phase actually needs to surface
Why developer selection is the decision no one takes seriously enough
Platform selection is a governance decision, not a technology one
Frontier safety
Why the safety conversation belongs in the transformation room
Frontier AI safety research and enterprise transformation practice are two communities not talking to each other.
Read →
Field observations
Agentic AI and the governance question no one is asking
Organisations are deploying agentic systems without asking who is accountable when the agent decides differently.
Read →
Implementation
Platform selection is a governance decision, not a technology one
The CIO should be in the room. Usually they are not. Here is what happens when they are not.
Read →
Frontier safety
What frontier AI researchers know that transformation directors do not
The gap between what is known in the research community and what is understood in the enterprise is larger than most realise.
Read →
Field observations
The facilitation failure that ends AI programmes before they start
If the person running the room cannot challenge the most senior person in it, the programme is already compromised.
Read →
Implementation
Why the discovery phase is where most enterprise AI projects actually fail
Not in the technology. Not in the budget. In the structured interviews that never happened.
Read →
Learn

The framework. The thinking. The free resources.

Start here. Everything on this page is free. When you are ready to go deeper, the Products page has structured programmes and tools built from the same work.

Free download

Start with the failure modes

A one-page reference guide to the five most common and expensive failure modes in agentic AI implementation. Built from live engagements across enterprise functions. No email required.

One-pager
The Agentic AI Implementation Framework — five phases, five failure modes
The framework

What the framework is and where it comes from

The Agentic AI Implementation Framework maps the five phases every enterprise AI programme moves through, and names the failure mode that ends most programmes at each one. It was built from live engagements across finance, procurement, legal, HR and controllership functions — not from theory or research alone.

The failure modes are not edge cases. They appear in almost every engagement, at every level of seniority, in organisations that are otherwise well-run. Naming them does not prevent them. Understanding them early, before the decisions get made, does.

Stay informed

Products and programmes in development

I am building structured learning resources — workshops, a mini-course, toolkits, cohorts. Leave your email and I will let you know when they launch. No noise.

Products

Tools and programmes built from practice, not theory

Every product here comes from 15 years of enterprise transformation and three years at the frontier of AI. Built for people who need to act, not just understand.

Assess
Understand where you are before deciding what to do next.
Facilitated session
Scorecard workshop
The scorecard run as a facilitated half-day with your leadership team. Outputs include a heatmap, prioritised roadmap and executive summary ready for board presentation.
From £2,000
Waitlist

Implement
Practical tools for organisations ready to move from principles to action.
Downloadable toolkit
AI policy-to-practice toolkit
Helps organisations move from responsible AI principles to practical implementation. Includes: AI use case intake template, risk-tiering model, governance RACI, approval workflow, model register template, vendor risk checklist, incident escalation process and executive steering committee pack.
£149 toolkit  |  £499 with implementation guide
Coming soon

Reference guide
Five failure modes one-pager
The most common and expensive failure modes in agentic AI implementation. Built from live engagements. Free to download from the Learn page.
Free
Available now

Learn
Structured learning for transformation professionals working through this in real time.
Live session
90-minute implementation workshop
Walk through the five-phase framework and apply it to your specific function and context. For individuals or small groups.
£500 per person
Waitlist

Self-guided
3-part mini-course
Work through the framework at your own pace. Video, frameworks and templates included. Built for individual professionals paying personally.
TBC
Waitlist

On demand
Recorded training module
The full framework on demand. Watch, pause, apply. Revisit when you hit each phase in a live engagement.
TBC
Waitlist

Cohort
Peer learning cohort
A small group of transformation directors working through implementation together, with facilitation. Shared context, real problems.
TBC
Waitlist

1-to-1 or small group
Work through a challenge with me
You bring the problem. I bring 15 years of enterprise transformation and a frontier safety lens. Structured conversation, written output.
Book to discuss
Available now

More in development.
Join any waitlist
to stay informed.
For senior leaders
Executive programmes for boards and leadership teams who need to lead in an AI-shaped organisation.
Executive briefing
Half-day AI briefing for leadership teams
Not how to use AI tools. How to lead, govern and make decisions in an AI-shaped organisation. Covers AI strategy, responsible AI, workforce implications, operating model redesign and what boards should ask before approving AI investment.
Enquire for pricing
Coming soon

Career transition
Moving into AI-adjacent roles
For experienced corporate professionals moving into AI product management, transformation leadership, responsible AI, governance or strategy. Not becoming an engineer. Becoming the person who leads.
6-week cohort, pricing TBC
Interest list

Advisory
Executive AI advisory
Premium advisory for boards and executive teams who need to understand AI well enough to make decisions, challenge hype and govern risk. Bridging boardroom and delivery reality.
By referral only
Future offering

All products reflect Anca Iftode's independent work. Views and frameworks are her own and do not represent any employer or organisation.
Speaking

I speak on frontier AI and enterprise reality

Conferences, panels, podcasts, internal knowledge sessions and professional networks. The safety lens throughout.

Formats
Keynote
Panel discussion
Fireside chat
Webinar
Podcast guest
Lunch and learn
Facilitated learning session
Topics
AI implementation beyond the hype
What makes enterprise AI transformation fail, and what the research says about why. Practical, honest, no vendor agenda.
Why frontier safety matters for business adoption
What safety researchers know that transformation directors do not and should. Bridging two communities that are not talking to each other.
From pilots to operating models
The failure modes between proof of concept and production. Why pilots succeed and transformations still fail.
Agentic AI and the future of work
What agentic systems mean for accountability, governance and the human in the loop.
The gap between AI labs, policy and the enterprise floor
Why the conversation happening in research is not reaching the people who need it most.
Work with me

Bring the problem. I bring the lens.

Two ways to work with me. Both are structured, time-bounded and outcome-focused.

Professionals and teams

Work through a challenge with me

You are implementing agentic AI inside a large function and something is not working. Maybe it is the framing, the platform decision, the developer brief, or the room dynamics. I work with individuals and small groups — one to one or up to five people — to work through the specific challenge in front of you.

This is a structured conversation, not a consulting engagement. You bring the problem. I bring 15 years of enterprise transformation and a frontier safety lens.

Who this is for
Transformation directors, senior programme leads, enterprise function heads implementing agentic AI
Format
One to one or up to five people, video call
Duration
90 minutes
What you get
A structured working session, a written summary, and a set of prioritised next actions
Small businesses

AI audit for owner-managed businesses

Fixed-price AI audit for owner-managed businesses of 5 to 50 people in Wandsworth and South West London. I look at your current processes, identify where AI can make a real difference, and give you a clear written brief on what to do and in what order. No jargon, no platform agenda, no retainer.

Who this is for
Owner-managed businesses, 5 to 50 people, Wandsworth and SW London
What you get
Written audit brief with prioritised recommendations and implementation roadmap
Pricing
Fixed price, agreed before we start, no surprises
First step
Free 20-minute discovery call to confirm fit
Speaking and learning sessions

Invite me to speak or book a session

For speaking invitations, podcast enquiries, panel discussions and learning sessions for professionals and teams.

Small business enquiries

Book a free discovery call

For owner-managed businesses in Wandsworth and South West London interested in the AI audit.