# AI for software and IT companies: which tasks to automate first Updated October 2026 Across the 3.6 million US jobs in software and IT services, the work scores 58 of 100 for AI exposure, weighted by how many people do each job. AI can take on about 918 hours a year per full-time person, 46% of a working year. The task that frees the most hours across the industry 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 918 hours a year per full-time person Task time AI can do today 18% AI speeds up another 72% ## Tasks to automate first The 15 tasks where AI can take on the most hours across software and IT services: 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, software testers, computer hardware engineers. | 68% AI can do it 97 hours a year for each person in the role | | Modify existing software to correct errors, adapt it to new hardware, or upgrade interfaces and improve performance. Coding assistants draft most fixes; developer still verifies hardware and integration. Done by software developers. | 65% AI helps 97 hours a year for each person in the role | | Analyze information to determine, recommend, and plan installation of a new system or modification of an existing system. AI can analyze requirements and draft installation plans; humans decide on changes. Done by software developers. | 50% AI helps 72 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. - **18%** AI can do it - **72%** AI helps - **10%** Needs a person - Supervise the work of programmers, technologists and technicians and other engineering and scientific personnel. Software developers, 15% exposed - Supervise and assign work to programmers, designers, technologists, technicians, or other engineering or scientific personnel. Software developers, 14% exposed - Set up equipment for employee use, performing or ensuring proper installation of cables, operating systems, or appropriate software. IT support specialists, 15% exposed - Inspect equipment and read order sheets to prepare for delivery to users. IT support specialists, 15% exposed - Distribute promotional materials at meetings, conferences, or trade shows. Sales representatives of services, 8% 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, support and sales staff use approved assistants, and code and customer data stay 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 Ticket triage, release notes, test generation and CRM updates 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 answers product questions from your docs and opens a bug report with the details when it finds one. 4 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/) ## Teams in this industry The business functions in software and IT services, by share of the workforce, with each one's exposure. 1. [IT and software, 52.6% of the workforce, score **58**](https://www.aigentcy.com/ai-for/it/) 2. [Operations and admin, 13% of the workforce, score **52**](https://www.aigentcy.com/ai-for/operations/) 3. [Sales, 9.9% of the workforce, score **61**](https://www.aigentcy.com/ai-for/sales/) 4. [Marketing and content, 6.2% of the workforce, score **61**](https://www.aigentcy.com/ai-for/marketing/) 5. [Finance and accounting, 4.4% of the workforce, score **59**](https://www.aigentcy.com/ai-for/finance/) ## Roles in software and IT services The 20 largest roles, each linked to its task-by-task score. Shares come from the BLS industry staffing estimates, May 2025. Of the roles with at least 2% of the jobs, customer service representatives score highest at 74, general managers lowest at 44. 1. [Software developers, 21.7% of the workforce, score **59** out of 100](https://www.aigentcy.com/ai-exposure/software-developers/) 2. [IT support specialists, 6% of the workforce, score **59** out of 100](https://www.aigentcy.com/ai-exposure/it-support-specialists/) 3. [Sales representatives of services, 5.5% of the workforce, score **66** out of 100](https://www.aigentcy.com/ai-exposure/sales-representatives-of-services/) 4. [IT managers, 5.5% of the workforce, score **50** out of 100](https://www.aigentcy.com/ai-exposure/it-managers/) 5. [Systems analysts, 4.4% of the workforce, score **59** out of 100](https://www.aigentcy.com/ai-exposure/systems-analysts/) 6. [General managers, 3.8% of the workforce, score **44** out of 100](https://www.aigentcy.com/ai-exposure/general-managers/) 7. [Project managers, 3% of the workforce, score **53** out of 100](https://www.aigentcy.com/ai-exposure/project-managers/) 8. [Customer service representatives, 2.7% of the workforce, score **74** out of 100](https://www.aigentcy.com/ai-exposure/customer-service-representatives/) 9. [Digital marketers, 2.4% of the workforce, score **62** out of 100](https://www.aigentcy.com/ai-exposure/digital-marketers/) 10. [Software testers, 2.3% of the workforce, score **64** out of 100](https://www.aigentcy.com/ai-exposure/software-testers/) 11. [Management consultants, 2% of the workforce, score **55** out of 100](https://www.aigentcy.com/ai-exposure/management-consultants/) 12. [Systems administrators, 1.8% of the workforce, score **58** out of 100](https://www.aigentcy.com/ai-exposure/systems-administrators/) 13. [Sales managers, 1.8% of the workforce, score **45** out of 100](https://www.aigentcy.com/ai-exposure/sales-managers/) 14. [Computer network architects, 1.6% of the workforce, score **55** out of 100](https://www.aigentcy.com/ai-exposure/computer-network-architects/) 15. [Cybersecurity analysts, 1.6% of the workforce, score **60** out of 100](https://www.aigentcy.com/ai-exposure/cybersecurity-analysts/) 16. [Accountants, 1.5% of the workforce, score **60** out of 100](https://www.aigentcy.com/ai-exposure/accountants/) 17. [Data scientists, 1.4% of the workforce, score **67** out of 100](https://www.aigentcy.com/ai-exposure/data-scientists/) 18. [Marketing managers, 1.3% of the workforce, score **55** out of 100](https://www.aigentcy.com/ai-exposure/marketing-managers/) 19. [Recruiters, 1.3% of the workforce, score **62** out of 100](https://www.aigentcy.com/ai-exposure/recruiters/) 20. [Network engineers, 1.2% of the workforce, score **59** out of 100](https://www.aigentcy.com/ai-exposure/computer-network-support-specialists/) ## Questions about AI for software and IT companies Which tasks should a software company 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 industry. How many hours can AI save a software company? About 918 hours a year per full-time person, 46% of a 2,000-hour working year. For a team of 10 that is about 9,180 hours, the time of 4.6 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 a software company are most exposed to AI? Of the roles with at least 2% of the jobs, customer service representatives score highest, at 74 of 100, and general managers lowest, at 44. Each role's page lists its tasks one by one. Will AI replace jobs in software and IT services? AI can do 18% of the task time across these roles today. Another 72% 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 a software company start with AI? With the quick wins: approved assistants for the tasks AI helps with, under a written AI policy. Then automate the 4 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/) ### Find the tasks to automate first in your team We map the tasks in your business, 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.