AI Threat Assessment · 26 May 2026

Deloitte

Big Four Consulting
COOKED
6.3/ 10

“Deloitte spent 175 years perfecting the art of selling human hours at a premium — armies of MBAs armed with PowerPoint templates, process maps, and the unshakeable confidence that comes from charging $300/hour for work a senior associate actually does. The exquisite irony is that while they're busy helping Fortune 500 companies 'transform digitally,' they've built a business model that's essentially a human vending machine in an era when Claude can draft the same strategic recommendations, McKinsey frameworks, and implementation roadmaps without the billable hour markup or the two-week delay for 'additional research.'”

Business Model
7.5
Automation Risk
7.0
Moat Strength
3.5
Adaptability
6.0
Need Survival
5.5
AI Threat Level
BUSINESS MODEL REPLACEABILITY

Their value proposition boils down to 'we have seen this problem before and can pattern-match a solution' — which was genuinely valuable until GPT-4 ingested every business case study ever written and started offering the same insights without the $50K minimum engagement fee. The billable hour model survives exactly until clients realize they're paying Mercedes prices for Honda Civic solutions.

7.5
WORKFORCE AUTOMATION RISK

Junior analysts doing research, associates building slide decks, and managers writing executive summaries are already living on borrowed time — Claude produces the same deliverables in minutes, not weeks. The partners who golf with C-suites are safe; everyone below director level is essentially a very expensive human wrapper around work that AI now does natively.

7.0
MOAT STRENGTH

The genuine moat isn't the consulting methodology — it's the Rolodex. Twenty-year relationships with Fortune 500 CFOs, regulatory expertise that takes decades to build, and the liability insurance to sign off on billion-dollar implementations. The work gets commoditized; the signature authority and the CEO's golf buddy status remain scarce.

3.5
AI ADAPTABILITY SIGNALS

They've launched an 'AI & Engineering' practice and mention generative AI throughout their site, but the adaptation feels like strapping a Tesla engine onto a horse-drawn carriage — impressive marketing, fundamentally unchanged delivery model. They're still selling human-hour packages while pitching AI transformation to clients who could just buy Claude directly.

6.0
WILL THE NEED SURVIVE AI?

Strategic advice survives; the 47-slide deck explaining why you need strategic advice does not. The core question shifts from 'what should we do?' to 'who do we trust to help us execute it?' — and execution increasingly means supervising AI systems, not managing armies of consultants.

5.5
Verdict

“Deloitte isn't getting killed by AI — they're getting killed by the gap between what they charge for thinking and what thinking now costs. The partners with the client relationships survive; the pyramid of junior talent that generates the actual margin becomes the world's most expensive training dataset for work that Claude already learned.”

Scores are based on public information and AI analysis. This is an affectionate roast, not a financial assessment. The best companies use this as a mirror, not a verdict.

Roast another →