AVGPAY AI & WORK RESEARCH
How could AI change the work of software developer?
The evidence points to very high modeled task exposure (66/100) for this occupation. This does not mean a 66% chance of losing a job. Compare what the research can actually measure with typical pay, published demand and the work people do.
- AI task exposure
- 66 / 100 Very high · 2023 study assumptions
- National median pay
- $135,980 OEWS published wages
- Employment change
- +10.2% BLS 2025–2035 projection
- Average annual openings
- 95,300 Includes replacement openings
How this occupation compares
Software Developer has a score of 66, close to the computers & math field average of 65. Its score is higher than occupations employing approximately 96% of workers in the covered wage dataset. These are job-weighted comparisons, not a survey of worker AI usage.
- Tasks potentially accelerated by a chatbot alone
- 42%
- With supporting software and tools
- 90%
The model assumes some tasks could be completed at least twice as quickly with comparable quality. It does not say employers have automated those tasks or eliminated equivalent numbers of jobs. The two scenarios are modeled capabilities from the study's historical baseline, not measurements taken in 2026.
What work could change—and what still needs investigation?
Research, design, and develop computer and network software or specialized utility programs. Analyze user needs and develop software solutions, applying principles and techniques of computer science, engineering, and mathematical analysis. Update software or enhance existing software capabilities. May work with computer hardware engineers to integrate hardware and software systems, and develop specifications and performance requirements. May maintain databases within an application area, working individually or coordinating database development as part of a team. The activities below are documented occupation tasks, not a list of tasks that AvgPay has individually proven AI can automate.
- Analyze information to determine, recommend, and plan installation of a new system or modification of an existing system.Which information-handling steps could AI assist with, and which need human verification?
- Analyze user needs and software requirements to determine feasibility of design within time and cost constraints.Which information-handling steps could AI assist with, and which need human verification?
- Confer with data processing or project managers to obtain information on limitations or capabilities for data processing projects.Which information-handling steps could AI assist with, and which need human verification?
- Confer with systems analysts, engineers, programmers and others to design systems and to obtain information on project limitations and capabilities, performance requirements and interfaces.Which information-handling steps could AI assist with, and which need human verification?
- Consult with customers or other departments on project status, proposals, or technical issues, such as software system design or maintenance.Which information-handling steps could AI assist with, and which need human verification?
- Design, develop and modify software systems, using scientific analysis and mathematical models to predict and measure outcomes and consequences of design.Which information-handling steps could AI assist with, and which need human verification?
- Determine system performance standards.What judgment, interaction or domain expertise is necessary to complete this task well?
- Develop or direct software system testing or validation procedures, programming, or documentation.Which information-handling steps could AI assist with, and which need human verification?
Three practical questions help turn a general exposure score into a meaningful discussion: which tasks involve routine information processing; which need confidential data, legal responsibility or face-to-face judgment; and which parts of the workflow have actually adopted AI in your industry? The study cannot answer those for your employer.
Skills and preparation that shape the decision
These are important occupational skills reported in the skills inventory. They are not evidence of immunity to automation or recommendations to abandon existing expertise.
- Critical Thinking
- Active Learning
- Reading Comprehension
- Active Listening
- Writing
Typical entry education: Bachelor's degree. Related experience: None. On-the-job training: None. A transition may require credentials, licenses and specialized experience even when two jobs appear similar.
BLS projects an increase of 10.2% in occupation-level employment between 2025 and 2035, from approximately 1,717,800 to 1,892,600 positions. Its annual openings measure includes people leaving an occupation, which explains why openings can exist even with negative net growth. The projection is not a causal estimate of AI job displacement.
So, is this job likely to disappear?
That cannot be determined from the current exposure score. Exposure measures theoretical task acceleration, not probability of replacement, employment effects, business investment or the pace at which a profession adapts. The best decision is to compare this occupation's current tasks with the latest BLS growth outlook, then investigate actual AI adoption where you work.
This page includes a source-backed task inventory, occupational outlook and direct exposure study match.
Full salary, forecast scenarios and career paths → · Search another occupation → · Workforce research report →
Methods: AvgPay exposure methodology. Original study: Eloundou, T., Manning, S., Mishkin, P. and Rock, D. GPTs are GPTs: Labor market impact potential of LLMs. Science (2024); working paper arXiv:2303.10130 (2023).. Employment and salary sources are documented in each career profile.