Short answer
Very unlikely. BLS rates welders among the occupations least exposed to AI, and most of the work is the kind current AI tools cannot do.
AI and jobs
The Bureau of Labor Statistics places welders in its low AI exposure category, among the occupations least exposed to current AI. BLS projects employment to change by +2.4% over 2025–2035, against 3.5% for all occupations, with about 40,300 openings a year.
Short answer
Very unlikely. BLS rates welders among the occupations least exposed to AI, and most of the work is the kind current AI tools cannot do.
The federal measure
The Bureau of Labor Statistics publishes a relative AI exposure category for each occupation it projects, built from five public measures of where AI could be, or already is, used to assist with or complete the work. Welders land in the low category: among the occupations least exposed to current AI.
Where this job sits on the BLS scale
786 occupations in the NueCareer bank, in BLS's four relative categories; width shows how many fall in each
Source: BLS Employment Projections, AI exposure categories supplemental table.
Inside the work
O*NET rates how important each work activity is to an occupation and how often each working condition comes up. The ratings below are the ones that move the job's NueCareer AI-resilience score the most: the ones that pull it up, and the ones that pull it down. Only ratings clearly above the midpoint count.
Physical, hands-on and person-to-person work
Routine, repeatable and screen-based work
O*NET 31.0 Database task statements, most important first.
The jobs
Exposure says where AI meets the work. The BLS projection says what the labor market is expected to do over 2025–2035. BLS projects +2.4% change in employment over the decade, against 3.5% across all occupations, and about 40,300 openings a year once retirements and career changes are counted. About 416,210 people work in the job today, at a median of $53,750 a year (BLS OEWS, May 2025).
Options
Careers O*NET lists as related to the work, or in the same field at similar pay, that BLS places in a low exposure category.
Installation, Maintenance, and Repair
Repair, maintain, or install electric motors, wiring, or switches.
Production
Construct, assemble, or rebuild machines, such as engines, turbines, and similar equipment used in such industries as construction, extraction, textiles, and paper manufacturing.
Production
Fabricate, position, align, and fit parts of structural metal products.
Set up, operate, or tend welding, soldering, or brazing machines or robots that weld, braze, solder, or heat treat metal products, components, or assemblies.
Construction and Extraction
Construct, assemble, maintain, and repair stationary steam boilers and boiler house auxiliaries.
Set up, operate, or tend woodworking machines, such as drill presses, lathes, shapers, routers, sanders, planers, and wood nailing machines.
Questions
Very unlikely. BLS rates welders among the occupations least exposed to AI, and most of the work is the kind current AI tools cannot do. BLS defines exposure as AI being able to assist with or complete some of the work, and says it does not imply job loss, productivity gains or wage effects. BLS projects employment to change by +2.4% over 2025–2035, against 3.5% for all occupations, with about 40,300 openings a year.
BLS sorts 786 occupations into four relative categories: low, moderate, high and very high. Welders are in the low category, alongside 202 other occupations. The categories combine five public measures of where AI could be or already is used, and they compare occupations with each other rather than giving an absolute risk.
In O*NET's description of the work, the parts that overlap most with what AI handles well are getting information, being exact or highly accurate and repetitive motions. The parts that hold up best are face-to-face discussions, hands-on work with tools and objects and handling and moving objects. These come from O*NET ratings of how important each activity is to the job, read through NueCareer's AI-resilience method.
On NueCareer's AI-resilience score, built from O*NET work activities, welders score 45 out of 100, higher than 51% of the occupations we score. The job pays a median of $53,750 a year (BLS OEWS, May 2025). No job is fully safe from technological change, but the federal exposure category and the projected openings are the two numbers worth weighing.
Electric Motor, Power Tool, and Related Repairer (low exposure, $56,210 median), Engine and Other Machine Assembler (low exposure, $53,710 median), Structural Metal Fabricator and Fitter (low exposure, $51,330 median), Welding, Soldering, and Brazing Machine Setter, Operator, and Tender (low exposure, $47,920 median). Each is also in the low category, and most are listed by O*NET as related to the work.
The AI exposure category is the Bureau of Labor Statistics'. The AI-resilience score and the activity breakdown are NueCareer's analysis of O*NET 31.0 Database data (method), not a federal figure. More: jobs most exposed to AI, jobs least exposed to AI, or check any job.