Lion Transformation Partners
Insights & Research

The evidence is clear.
The window is closing.

Our point of view is built on research from McKinsey, Harvard, MIT, Deloitte, WEF, BCG, EY, PwC, OECD, IMF, and Forrester. Here is what the data actually says — and what it means for your organisation.

The Burning Platform

AI investment is universal.
ROI is not.

The gap is not a technology problem. It is an operating model problem. Companies that win are not using better AI — they are deploying it with the infrastructure to make it stick.

95%

of GenAI projects fail to show measurable ROI

MIT 2025
79%

of organisations struggle to adopt AI at scale

Writer 2026
54%

of C-Suite say AI adoption is tearing their org apart

Writer 2026
6%

qualify as AI high performers with 5%+ EBIT impact

Fullview 2025

“Companies winning with AI are not using better models. They are deploying, adopting and scaling faster. Competitive distance compounds — and becomes irreversible.”

— McKinsey, April 2026
Why Now

Three forces are converging.
The window to lead is 18 months.

Miss this window and you are catching up — not competing. Competitive distance compounds and becomes irreversible.

Agentic AI Is Operational

AI agents now orchestrate multi-step enterprise workflows without human involvement. Organisations building AI-native operating models today have an 18-month head start. That lead compounds.

90% of Claude Code written by Claude itself

Allie Miller, Feb 2026

Regulation Arrives August 2026

EU AI Act high-risk obligations go live August 2026. Governance cannot be retrofitted. It must be built in from day one.

Fines up to €35M or 7% of global annual turnover

EU AI Act, 2026

Boards Demand ROI

CEOs, boards, and investors demanding demonstrable ROI on AI. The shift from ‘Is AI smart?’ to ‘Is AI valuable?’ is complete.

Only 29% see significant ROI despite individual gains

Writer 2026
The Human-AI Gap

Three types of AI user.
Only one creates value.

You can have 40,000 employees using AI daily — and still be wasting millions if they're only rewriting emails. The gap between adoption and impact is where transformation lives.

~5%Superusers
  • Move from chat to agents to systems
  • Run 10–40 concurrent AI tasks simultaneously
  • Treat AI as a teammate, not a tool
  • Measure economic impact, not usage frequency
~55%Surface Users
  • Use AI daily — but only rewrite emails
  • Summarise meetings, generate first drafts
  • Believe they are AI-savvy — they are not
  • Single-threaded, reactive, prompt-dependent
~40%Non-Users
  • Rarely or never use AI tools
  • Intimidated by interface complexity
  • Waiting for training that never comes
  • Risk of workforce irrelevance accelerating

The LTP perspective:

The largest group — surface users — holds enormous latent value. Converting them into orchestrators requires structured protocols, context systems, and human-centered change management. Not more training days. Infrastructure.

The Future of Work

When AI does 40–70% of your job,
what does the human do?

Research across McKinsey, MIT, Harvard, EY, WEF, and the IMF converges on six functions that are genuinely difficult to replicate — not just romantically comforting.

01

Orchestrating Intelligence

Directing entire symphonies of AI agents working in parallel. Defining what agents are for, what success looks like, and when to intervene.

02

Irreducible Human Judgment

Reading situations AI has never been in. Calling out technically correct answers that are ethically wrong. Deciding under genuine uncertainty.

03

Holding Relationships & Trust

Trust between organisations, teams, and leaders is an inherently human economy. AI can inform it. It cannot create it.

04

Providing Meaning & Purpose

Answering why this work matters. Protecting dignity during transformation. AI cannot be the meaning-maker.

05

Governance, Ethics & Accountability

Every consequential AI decision needs a human who can be held accountable — legally, morally, and strategically.

06

Creative Synthesis & Innovation

Cognitive leaps and genuinely novel breakthroughs still require humans in the loop. AI recombines patterns. Humans generate new questions.

170M

new roles created by 2030

WEF Future of Jobs 2025
39%

of workforce skills obsolete by 2030

WEF · McKinsey · PwC 2025
30%

of work hours automated by 2030

McKinsey New Future of Work 2025

“AI won't replace humans — but humans with AI will replace humans without AI.”

— Karim Lakhani, Harvard Business School

The human is the moat.

Technology is arriving faster than the human infrastructure to absorb it. AI brain fry, broken career ladders, and identity crises are change management problems — not tech problems.

An agent can draft the transition plan. It cannot hold the room when a team is afraid.

Start the conversation