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.
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.
of GenAI projects fail to show measurable ROI
MIT 2025of organisations struggle to adopt AI at scale
Writer 2026of C-Suite say AI adoption is tearing their org apart
Writer 2026qualify 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 2026Three 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 2026Regulation 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, 2026Boards 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 2026Three 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.
- 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
- 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
- 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.
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.
Orchestrating Intelligence
Directing entire symphonies of AI agents working in parallel. Defining what agents are for, what success looks like, and when to intervene.
Irreducible Human Judgment
Reading situations AI has never been in. Calling out technically correct answers that are ethically wrong. Deciding under genuine uncertainty.
Holding Relationships & Trust
Trust between organisations, teams, and leaders is an inherently human economy. AI can inform it. It cannot create it.
Providing Meaning & Purpose
Answering why this work matters. Protecting dignity during transformation. AI cannot be the meaning-maker.
Governance, Ethics & Accountability
Every consequential AI decision needs a human who can be held accountable — legally, morally, and strategically.
Creative Synthesis & Innovation
Cognitive leaps and genuinely novel breakthroughs still require humans in the loop. AI recombines patterns. Humans generate new questions.
new roles created by 2030
WEF Future of Jobs 2025of workforce skills obsolete by 2030
WEF · McKinsey · PwC 2025of 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 SchoolThe 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