Software Quality Assurance Analysts and Testers
This occupation overlaps heavily with what AI already does.
AI is already used, successfully, for a large share of this occupation's core activities. That doesn't schedule a layoff; it schedules a redesign. The people who direct AI through these tasks will set the pace for everyone else.
Where this occupation sits among all 785 measured. Bands are quintiles of the real distribution. Applicability measures task overlap with observed AI usage. It is not a probability of job loss.
What this job involves
Develop and execute software tests to identify software problems and their causes. Test system modifications to prepare for implementation. Document software and application defects using a bug tracking system and report defects to software or web developers. Create and maintain databases of known defects. May participate in software design reviews to provide input on functional requirements, operational characteristics, product designs, and schedules.
Where AI overlaps this work, activity by activity
The occupation's most important work activities (O*NET weights, normalized to this set), each with its measured AI performance across 200,000 real conversations: how often AI is used for it, how well it completes it, and how much of the activity it covers.
| Work activity | Share of role | AI performance |
|---|---|---|
| Develop operational or technical procedures or standards. | 18.5% | High overlap 0.68 |
| Test performance of computer or information systems. | 16.4% | Moderate 0.48 |
| Advise others on the design or use of technologies. | 10.8% | High overlap 0.75 |
| Maintain operational records. | 9.6% | Not observed 0.00 |
| Monitor operation of computer or information technologies. | 7.3% | Moderate 0.47 |
| Set up computer systems, networks, or other information systems. | 6.3% | High overlap 0.63 |
| Resolve computer problems. | 5.7% | High overlap 0.64 |
| Analyze data to improve operations. | 5.7% | Not observed 0.00 |
| Compile records, documentation, or other data. | 5.7% | Moderate 0.55 |
| Notify others of emergencies or problems. | 5.7% | Not observed 0.00 |
| Document technical designs, procedures, or activities. | 4.3% | High overlap 0.61 |
| Design databases. | 3.9% | Not observed 0.00 |
"Not observed" is a measurement, not missing data: that activity does not meaningfully appear in the usage sample. AI performance = completion × scope × coverage (user-goal view), 0–1.
What drives this score: the biggest AI-overlapped activity in this role is develop operational or technical procedures or standards; the most important activity AI is not observed doing is maintain operational records. The plan below is built around exactly that split.
High exposure playbook: own the tools, own the sign-off.
Next 30 days
- Adopt before you're outpaced. Pick your highest-overlap activity, Develop operational or technical procedures or standards, and run it through a current AI tool this week. Learn where it's genuinely good and exactly where it breaks.
- Start an error log. Every AI mistake you catch is evidence for the reviewer role this occupation is drifting toward, and your case for owning it.
- Audit your week. Split your tasks into 'AI can draft it' vs 'a human must own it'. The second list is your future job description.
Next 90 days
- Become the sign-off. Volunteer to define how AI output gets checked in your team: quality bars, review steps, accountability. The person who writes the checklist doesn't get replaced by it.
- Shift visible value to the resilient column. Push more of your time into work like Maintain operational records, the activities AI is not observed doing, and make sure your manager sees the shift.
- Ship one workflow. Combine AI plus your judgment into a repeatable process that makes you measurably faster than peers. Speed with quality is the currency now.
The longer game
- Reposition your title around judgment. In Computer & Mathematical, the durable roles concentrate direction, review, client trust and accountability. Move your CV language, and your actual duties, there before the market forces it.
- Check the adjacent moves below. Several nearby roles share most of your activities with materially lower exposure; a sideways step early beats a forced one later.
- Keep receipts. Document outcomes you drove that AI couldn't have: the negotiation saved, the error caught, the client kept. That portfolio is your moat.
- AI applicability score and per-activity metrics: Tomlinson, Jaffe, Wang, Counts & Suri, Working with AI: Measuring the Applicability of Generative AI to Occupations, Microsoft Research 2025 (arXiv:2507.07935), data CC BY 4.0.
- Occupation description and activity structure: O*NET 30.0 Database, USDOL/ETA, CC BY 4.0.
- Bands, personalization weights and action plans are JobRiskAI editorial layers; method and limitations on the methodology page.