The C-Suite Admits the Strategy Was Theatre
Fifty-four percent of CEOs now say AI is tearing their company apart. Seventy-five percent say the strategy is for show. The confession is the signal.
For three years, the question of whether AI transformation was working has been answered mostly by vendors, consultants, and the CEOs trying to sell the narrative to their own boards. This week, the C-suite admitted the strategy is a performance. Stanford confirmed that the number reAImagine.work put on early-career displacement in Issue 001 has gotten worse. And Baker McKenzie became the first global law firm to translate AI into mass layoffs. Three signals, one direction: the performance is over, the accounting has begun.
Signal 1: The C-Suite Finally Admits the Strategy Was Theatre
WRITER and Workplace Intelligence surveyed 1,200 C-suite executives and 1,200 employees for the 2026 AI Adoption in the Enterprise report, released this month. The findings are not subtle. Fifty-four percent of executives now say AI adoption is "tearing their company apart." Seventy-five percent admit their AI strategy is "more for show" than actual internal guidance. Forty-eight percent call AI adoption "a massive disappointment," up from 34 percent a year ago. Only 29 percent report significant ROI from generative AI. Only 23 percent from AI agents. Sixty-nine percent are planning layoffs attributed to AI, but 39 percent have no formal strategy for driving revenue from those tools.
Read the last two numbers together. Two-thirds of large companies are firing people in the name of an AI strategy that four in ten cannot describe as a revenue plan. This is the admission that consultants, vendors, and transformation leaders have been working for three years to prevent reaching the board. It has now reached the board, and the board is telling the surveyors the truth.
The structural cause is visible in the survey itself. Ninety-two percent of executives say they are actively cultivating "AI elite" employees while 60 percent plan layoffs for non-adopters. Sixty-seven percent believe their company has already suffered a data breach from unapproved AI tools. This is what happens when AI is deployed as a procurement decision rather than an operating redesign. Productivity gains concentrate with a few individuals. The rest of the organisation fractures around them.
Why this matters for your work: If your board is still being shown green dashboards and change-management velocity charts, you are in the 75 percent. The honest question for this week is not "how is our AI transformation progressing." It is "what would we say if we were answering an anonymous survey."
Signal 2: Stanford Confirms It. The Entry-Level Door Is Closing Faster Than We Said
In Issue 001, reAImagine.work published a 14 percent decline in hiring among young workers in AI-exposed roles as the most important workforce number of the quarter. The 2026 Stanford AI Index, released this month, has confirmed the pattern and updated the number. Employment among software developers aged 22 to 25 has now dropped nearly 20 percent since 2024, even as headcount among older developers in the same companies has grown. The same pattern appears in customer service. Stanford's Digital Economy Lab, drawing on ADP payroll data across millions of US workers, found the decline concentrated in occupations most exposed to generative AI and not explained by interest rates, post-pandemic normalisation, or firm-level shocks.
The shift in the number matters less than the shift in the shape. It is no longer a partial signal visible in one dataset. It is confirmed across ADP payroll records, Lightcast job-posting data, and the Anthropic Economic Index. Agentic AI skills moved from 0.06 percent to 0.23 percent of US job postings in a single year, a 280 percent increase. The work that used to go to graduates is being absorbed by agents, and the work that used to feed the mid-career bench is not being created.
For India and the GCC, where Global Capability Centres have been the fastest-growing employer of graduate analytical talent, this is the signal to watch. The ADP data is American. The model of work being displaced is global. When Lightcast reports that Python, AWS, and workflow management are the fastest-growing AI skills, it is describing exactly the work GCCs in Bengaluru and Riyadh were built to do.
Why this matters for your work: Organisations that quietly stopped graduate hiring in 2024 because "AI can do the work" are now staring at the 2027 promotion pipeline. There is nothing in it. Rebuilding the graduate intake takes three years. The firms that move this quarter will have bench depth in 2029. The firms that wait will not.
Signal 3: Baker McKenzie Opens the Professional Services Front
Global law firm Baker McKenzie announced in February that it is cutting between 600 and 1,000 employees, up to 10 percent of its global workforce, as part of a shift towards AI. The cuts are not paralegals and administrative staff, although they are included. The cuts reach into associate and business services ranks. It is the first time a global elite law firm has attached an AI label to a mass layoff, and it follows Allen & Overy's merger-era restructuring and Chegg's 45 percent workforce reduction in legal-adjacent education services.
Law has been the canonical "AI-resistant" profession in every consultant report since 2023. Partner-track work, the argument went, was protected by privilege, judgment, and client trust. The Baker McKenzie announcement does not contradict that argument. It sharpens it. What is being cut is the work below the partner track: document review, contract drafting, due diligence, research memos. The pyramid structure that made the traditional law firm economics work is being compressed from below. The same compression is now visible at professional services adjacencies. Deloitte, PwC, EY, and KPMG have all reduced graduate intake in their consulting and audit arms during the past twelve months, though none have attached an AI label to the reductions. Baker McKenzie has made the label public.
For clients in India, the GCC, and Africa, the downstream effect arrives within two quarters. Billable-hour models in legal and consulting engagements are being rebuilt around AI-assisted delivery. The price negotiation coming to your next Magic Circle or Big Four engagement will not be about rate cards. It will be about what share of the savings from AI-assisted delivery is returned to you versus retained by the firm. In India specifically, the domestic legal and consulting market has long priced on Indian rates regardless of AI usage. That will not hold through 2026.
Why this matters for your work: If your organisation procures legal or consulting services on a headcount-plus-margin basis, your procurement model is about to be obsolete. The firms cutting now will come back with AI-blended pricing before the end of the year. The buyers ready with the right questions will capture the savings. The buyers who are not will pay for the transition.
The three signals describe one arc. The admission, the confirmation, the conversion. In this week's Conversation section, I write about why organisations are doing AI-led transformation wrong, and what it would take to get it right. The WRITER survey numbers above are the evidence. The essay is the diagnosis.
Build acumen on this topic
The C-Suite Admits the Strategy Was Theatre
Turn what you just read into verified expertise. Start with a $5 AI tutor session or earn a $15 verified badge.
Start Learning