# AI for IT teams: which tasks to automate first Updated October 2026 Across the 5.5 million US jobs in IT and software, the work scores 58 of 100 for AI exposure, weighted by how many people do each job. AI can take on about 892 hours a year per full-time person, 45% of a working year. The task that frees the most hours across the team is to prepare reports or correspondence concerning project specifications, about 114 hours a year for each person who does it. AI exposure, weighted by employment 58 out of 100 High exposure Hours AI can take on 892 hours a year per full-time person Task time AI can do today 12% AI speeds up another 78% ## Tasks to automate first The 15 tasks where AI can take on the most hours across IT and software: how exposed each task is, how much of the job it takes and how many people do that job. Hours are per person in the role that does the task. [How we score each task](https://www.aigentcy.com/ai-exposure/methodology/). | Task and who does it | AI exposure | | --- | --- | | Prepare reports or correspondence concerning project specifications, activities, or status. Status reports and specification write-ups are routine writing AI handles well. Done by software developers. | 78% AI can do it 114 hours a year for each person in the role | | Develop or direct software system testing or validation procedures, programming, or documentation. AI drafts test cases, validation scripts and docs; engineer directs and approves. Done by software developers. | 65% AI helps 101 hours a year for each person in the role | | Store, retrieve, and manipulate data for analysis of system capabilities and requirements. Querying, transforming and retrieving data is routine digital work AI scripts handle. Done by software developers. | 68% AI can do it 97 hours a year for each person in the role | | Maintain records of daily data communication transactions, problems and remedial actions taken, or installation activities. Logging transactions and remedial actions is largely templated and automatable. Done by IT support specialists. | 72% AI can do it 98 hours a year for each person in the role | | Answer user inquiries regarding computer software or hardware operation to resolve problems. LLM assistants answer most routine software and hardware questions. Done by IT support specialists. | 65% AI can do it 97 hours a year for each person in the role | ## What stays with people 10% of the task time across these roles needs a person, and the output of the rest still needs someone to check it. The tasks below take the most of that time. - **12%** AI can do it - **78%** AI helps - **10%** Needs a person - Supervise the work of programmers, technologists and technicians and other engineering and scientific personnel. Software developers, 15% exposed - Set up equipment for employee use, performing or ensuring proper installation of cables, operating systems, or appropriate software. IT support specialists, 15% exposed - Install and perform minor repairs to hardware, software, or peripheral equipment, following design or installation specifications. IT support specialists, 18% exposed - Supervise and assign work to programmers, designers, technologists, technicians, or other engineering or scientific personnel. Software developers, 14% exposed - Meet with department heads, managers, supervisors, vendors, and others, to solicit cooperation and resolve problems. IT managers, 10% exposed ## Adoption path Start with the tasks where AI helps a person, then automate the routine ones, then give whole workflows to agents with a person approving the output. 1. ### Quick wins Developers and support staff use approved coding and troubleshooting assistants, and code stays on infrastructure you control. Example: Develop or direct software system testing or validation procedures. AI helps with 65% of the time. [AI governance](https://www.aigentcy.com/services/ai-governance/) 2. ### Workflow automation Password resets, access requests, test generation and incident triage run as automations. Example: Prepare reports or correspondence concerning project specifications. AI can do 78% of it. [Process automation](https://www.aigentcy.com/services/process-automation/) 3. ### Agents An agent works the first line of the service desk and writes up each fix in your knowledge base. 8 of the 15 tasks on this page are ones AI can do end to end, the work an agent takes on. [Private AI](https://www.aigentcy.com/services/private-ai/) ## Roles in IT and software 27 roles, each linked to its task-by-task score. Of the roles with at least 2% of the jobs, software testers score highest at 64, IT managers lowest at 50. 1. [Software developers, 30.9% of the team, score **59** out of 100](https://www.aigentcy.com/ai-exposure/software-developers/) 2. [IT support specialists, 13.1% of the team, score **59** out of 100](https://www.aigentcy.com/ai-exposure/it-support-specialists/) 3. [IT managers, 12.3% of the team, score **50** out of 100](https://www.aigentcy.com/ai-exposure/it-managers/) 4. [Systems analysts, 9.5% of the team, score **59** out of 100](https://www.aigentcy.com/ai-exposure/systems-analysts/) 5. [Systems administrators, 5.8% of the team, score **58** out of 100](https://www.aigentcy.com/ai-exposure/systems-administrators/) 6. [Cybersecurity analysts, 3.5% of the team, score **60** out of 100](https://www.aigentcy.com/ai-exposure/cybersecurity-analysts/) 7. [Software testers, 3.4% of the team, score **64** out of 100](https://www.aigentcy.com/ai-exposure/software-testers/) 8. [Computer network architects, 3.3% of the team, score **55** out of 100](https://www.aigentcy.com/ai-exposure/computer-network-architects/) 9. [Network engineers, 2.7% of the team, score **59** out of 100](https://www.aigentcy.com/ai-exposure/computer-network-support-specialists/) 10. [Programmers, 1.7% of the team, score **66** out of 100](https://www.aigentcy.com/ai-exposure/programmers/) 11. [Computer hardware engineers, 1.4% of the team, score **51** out of 100](https://www.aigentcy.com/ai-exposure/computer-hardware-engineers/) 12. [Web developers, 1.3% of the team, score **66** out of 100](https://www.aigentcy.com/ai-exposure/web-developers/) 13. [Database administrators, 1.3% of the team, score **60** out of 100](https://www.aigentcy.com/ai-exposure/database-administrators/) 14. [Data engineers, 1.2% of the team, score **60** out of 100](https://www.aigentcy.com/ai-exposure/data-engineers/) 15. [Blockchain engineers, 0.9% of the team, score **58** out of 100](https://www.aigentcy.com/ai-exposure/blockchain-engineers/) 16. [Digital forensics analysts, 0.9% of the team, score **59** out of 100](https://www.aigentcy.com/ai-exposure/digital-forensics-analysts/) 17. [Document management specialists, 0.9% of the team, score **61** out of 100](https://www.aigentcy.com/ai-exposure/document-management-specialists/) 18. [GIS technicians, 0.9% of the team, score **62** out of 100](https://www.aigentcy.com/ai-exposure/gis-technicians/) 19. [Information security engineers, 0.9% of the team, score **60** out of 100](https://www.aigentcy.com/ai-exposure/information-security-engineers/) 20. [Information technology project managers, 0.9% of the team, score **56** out of 100](https://www.aigentcy.com/ai-exposure/information-technology-project-managers/) 21. [Penetration testers, 0.9% of the team, score **60** out of 100](https://www.aigentcy.com/ai-exposure/penetration-testers/) 22. [Systems engineers, 0.9% of the team, score **60** out of 100](https://www.aigentcy.com/ai-exposure/systems-engineers/) 23. [Web administrators, 0.9% of the team, score **62** out of 100](https://www.aigentcy.com/ai-exposure/web-administrators/) 24. [Computer scientists, 0.7% of the team, score **51** out of 100](https://www.aigentcy.com/ai-exposure/computer-scientists/) 25. [Data warehousing specialists, Counted under a related job in US data, score **68** out of 100](https://www.aigentcy.com/ai-exposure/data-warehousing-specialists/) 26. [Health informatics specialists, Counted under a related job in US data, score **58** out of 100](https://www.aigentcy.com/ai-exposure/health-informatics-specialists/) 27. [Telecommunications engineering specialists, Counted under a related job in US data, score **57** out of 100](https://www.aigentcy.com/ai-exposure/telecommunications-engineering-specialists/) ## Questions about AI for IT teams Which tasks should an IT team automate first? The first task to automate is to prepare reports or correspondence concerning project specifications (78% exposed). Next come the tasks to develop or direct software system testing or validation procedures (65%) and to store, retrieve, and manipulate data for analysis of system capabilities and requirements (68%). Together they free the most hours across the team. How many hours can AI save an IT team? About 892 hours a year per full-time person, 45% of a 2,000-hour working year. For a team of 10 that is about 8,920 hours, the time of 4.5 full-time people. The figure is the share of task time AI can remove today, weighted by how many people do each job. Which roles in an IT team are most exposed to AI? Of the roles with at least 2% of the jobs, software testers score highest, at 64 of 100, and IT managers lowest, at 50. Each role's page lists its tasks one by one. Will AI replace jobs in IT and software? AI can do 12% of the task time across these roles today. Another 78% gets faster with AI while a person stays in charge. 10% needs a person. One task that stays with people is to supervise the work of programmers. Where should an IT team start with AI? With the quick wins: approved assistants for the tasks AI helps with, under a written AI policy. Then automate the 8 tasks on this page AI can do end to end. An AI audit maps the tasks in your team and ranks them the same way this page does. ## Next steps - [Enterprise AI in Slack: one assistant for company knowledge](https://www.aigentcy.com/blog/enterprise-ai-in-slack/) - [Case study: shared code context for a development team](https://www.aigentcy.com/case-studies/agentic-second-brain/) ### Find the tasks to automate first in your team We map the tasks in your IT team, score each one the way this page does and build the automations that pay back first. [Book an AI audit](https://www.aigentcy.com/audit/) Occupation data from the O\*NET 30.3 Database by the U.S. Department of Labor, Employment and Training Administration (USDOL/ETA), used under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) and modified by Aigentcy. O\*NET® is a trademark of USDOL/ETA. Wage, employment and projection data from the U.S. Bureau of Labor Statistics. EU occupation names from ESCO, European Commission. AI usage signals from the Anthropic Economic Index (CC BY) and Microsoft Research (CC BY 4.0). [Methodology and sources](https://www.aigentcy.com/ai-exposure/methodology/). Scores © Aigentcy. Quote them with a link to the page.