AI Systems Architect: how AI changes this job over time
Our best estimates, shown as ranges and grades - not exact predictions.
Right now, AI can already do about 70% of the day-to-day work in this job, and by 20 years from now that could be around 95%. There are likely to be fewer of these jobs over time - very roughly 25-60% of the 2024 number, 20 years out. Getting your first job here is not that easy today, and it looks set to get harder. This job has no special protection, so the trick is to keep building skills AI cannot copy.
What we assume: AI keeps getting cheaper and better; robots arrive more slowly - small effect by ~2031, bigger by ~2036, widespread by the mid-2040s. "Number of jobs" means how many jobs there will be compared with 2024 (100% = the same). "How hard to get in" runs from A (easy) to E (very hard).
How a AI Systems Architect job changes over time
AI can already do a large part of this kind of work, and it is getting better at it quickly. There is no licence or rule that slows that down, so the number of these roles is likely to fall over time and getting in will get harder. The part that lasts is the human side - sharp judgement, working with people, and deep knowledge that goes beyond what a tool can copy.
AI will handle much of the routine output in these roles, and employers will expect you to use it fluently. Entry becomes harder if the main thing you offer is the work AI can already do.
The number of these jobs is likely to shrink noticeably as AI takes on more of the middle of the work. People who have moved towards judgement, people skills, and deep specialist knowledge will fare best.
This far ahead, no one knows how capable AI will become - and with no special protection, these roles are among the most exposed. Build skills you can carry into more than one kind of work, and expect to retrain at least once.
The honest bottom line: there are likely to be fewer of these jobs over time, and the easy-to-automate version of the role is already shrinking. What lasts is the part AI still cannot do well - real judgement, trust, and skills you have built up across more than one area. Stay flexible, keep learning, and aim for the human end of the work.
How to aim for a AI Systems Architect career
You're looking ahead at this job. By the time you join, AI will already do more of it - so aim for the part that will still need a person.
What pushes this score up
Sources: exposure dial - Anthropic labour market research (2026), observed real-world AI usage by occupation. Job-security category and forecast - OpenAI, "The AI Jobs Transition Framework" (Richmond, 2026, OpenAI Economic Research), CC BY 4.0, matched to "Computer Systems Engineers/Architects" (15-1299.08). Scorecard grades and verdicts are CourseMap editorial judgment - we show forecasts as forecasts and own our conclusions.