BCG's Henderson Institute analyzed 165 million US jobs and arrived at one conclusion that should sit on every CEO's desk this quarter: 50 to 55 percent of American jobs will be substantially reshaped by AI within three years. The word is reshaped, not eliminated. Indiana manufacturers and mid-market companies fall directly in that window. The question for CEOs right now: which roles at your company are in that 55 percent, and where do your people sit today?
What BCG Actually Found
The BCG study assessed what percentage of each role's tasks will be automatable within the study window across 165 million US jobs. The finding: roughly half of the American workforce sits in jobs where 40 percent or more of daily tasks will change because of AI.
BCG was explicit about one distinction worth slowing down on. The reshape happens at the task level, not the role level. A supply chain coordinator does not disappear. The work shifts from manual data reconciliation toward oversight of AI-generated logistics recommendations. The role continues. The daily activity changes substantially.
The report also flagged something CEOs should weigh more than the 55 percent number itself. Companies that cut headcount before understanding their internal AI adoption velocity face a second wave of costs. When AI deployment accelerates on the back end, they rehire into roles they eliminated three quarters earlier.
of US jobs will be substantially reshaped by AI within three years, according to BCG's analysis of 165 million jobs
Source: BCG Henderson Institute, 2026What It Looks Like When Internal AI Adoption Accelerates
Cloudflare posted $639.8 million in Q1 2026 revenue, a 34 percent year-over-year increase. In the same quarter, the company announced the elimination of 1,100 positions, roughly 20 percent of its workforce.
What made the Cloudflare announcement stand apart from the layoff stories of the past 18 months: the company disclosed that internal AI tool usage had grown 600 percent over the prior three months.
That sequence is the key data point. Revenue up 34 percent, internal AI usage up 6x, headcount down 20 percent. The positions that went away were ones where AI tools had absorbed enough of the task load that the math no longer supported the headcount. Cloudflare did not cut and then deploy. The deployment data informed the decision.
The Internal Adoption Sequence
Inside every company going through accelerated AI adoption, three things tend to happen in order. AI tools get deployed, usually in engineering or customer-facing support roles first. Productivity per employee rises faster than the initial estimates. Then structural decisions follow the productivity data.
The companies that work through this sequence in order end up ahead of where they started. The ones that cut first, before the adoption data is clear, face a more expensive path on the back end.
The companies that handle workforce reshaping well will reassign work first, then make structural decisions later. Cutting before the measurement is done is the expensive version.
What Indiana Manufacturers Need to Know Right Now
Indiana is not a bystander in this. The Central Indiana Corporate Partnership reports that AI-related job postings in Indiana have grown 440 percent since 2020. Governor Braun's IN AI initiative, announced in May 2026, targets AI readiness training for one million Indiana employees.
Hyndman Industrial in Fort Wayne has begun deploying AI tools in plant operations. It is not alone. Indiana's manufacturing base, with its density of production scheduling, supply chain coordination, and quality control work, sits in the categories BCG identifies as facing the highest task-automation pressure.
The planning distinction: reshape pressure is role-level, not company-level. One Indiana plant might have three roles where AI tools will absorb 40 percent or more of the current task load, and twelve roles largely unchanged. The headline number does not tell you which is which at your company.
The Step That Counts Before the Headline Numbers Do
The BCG timeline gives Indiana CEOs a three-year window. That window is already running.
The companies that work through this well will do one thing before making structural decisions: measure where their teams actually sit on the AI proficiency scale. A manufacturing company whose workforce sits at Level 1 or Level 2 on The 7 Levels of AI Proficiency is in a different planning position than one whose team sits at Level 4 or above.
That baseline is not complicated to establish. The 7 Levels of AI Proficiency assessment at assess.launchready.ai takes under 10 minutes per team member and places each person on the proficiency scale. The output tells you where AI reshaping pressure falls hardest relative to your team's current capability.
BCG's warning about cutting before measuring applies to Indiana companies the same way it applies to Cloudflare. The workforce decisions that age well are the ones made with proficiency data in hand.
Related reading: Level 4: Commander in The 7 Levels of AI Proficiency.
Frequently Asked Questions
What does AI reshaping a job actually mean in practice?
BCG defines it as 40 percent or more of the tasks in a role becoming automatable within the study window. The role does not disappear. The daily work changes enough that the people doing it need different skills or different amounts of time to produce the same output. A quality control inspector role shifts from manually checking every item to overseeing AI-flagged anomalies. The job title stays. The task mix changes substantially.
Which Indiana job categories face the highest reshaping pressure?
Based on BCG's task-analysis methodology and Indiana's industry mix, the categories facing the most pressure are production scheduling, supply chain coordination, customer service intake, and back-office administrative work. Skilled trades and direct client-relationship roles show lower automation pressure. The key variable is not the job title but the task composition: roles where most of the work involves data entry, pattern matching, or rule-based decisions face higher pressure than roles built around relationship management, physical judgment, or complex problem-solving.
How do I measure my team's AI readiness before making workforce decisions?
The 7 Levels of AI Proficiency assessment at assess.launchready.ai takes under 10 minutes per person and places each team member on the proficiency scale from Level 1 (AI Aware) to Level 7 (AI Orchestrator). The output gives you a baseline for each role, which tells you where reshape pressure falls against your team's actual current capability. That baseline is the input to a workforce planning conversation.
Sources
- BCG Henderson Institute. "AI Will Reshape More Jobs Than It Replaces." bcg.com
- TechCrunch. Cloudflare layoff and AI adoption disclosure. May 8, 2026. techcrunch.com
- Cloudflare Blog. "Building for the Future." blog.cloudflare.com
- Central Indiana Corporate Partnership. "Indiana's AI Imperative: Building the Nation's Most AI-Ready Economy." cicpindiana.com
- Indiana Governor's Office. "Governor Braun Announces IN AI to Grow Jobs and Wages Through Human-Centered AI." May 2026. in.gov
- WTHR News. "Indiana Governor Mike Braun on AI Initiative." wthr.com
Find your AI Proficiency level
The free 7 Levels assessment places you across seven stages of AI capability. Under ten minutes. Research-backed scoring.