Date Posted: Thursday, 16th July 2026
According to much of the coverage around generative AI, your job is next. Software developers will be replaced by coding agents. Support teams will disappear behind chatbots. Analysts, testers and project managers will be automated out of existence.
The reality is considerably less dramatic. AI is unlikely to eliminate entire categories of technology professionals overnight. It is far more likely to automate specific tasks, reshape team structures and change what organisations expect from the people they employ. AI is not so much wiping jobs out as evolving them.
The International Labour Organisation estimates that one in four workers globally is in an occupation with some exposure to generative AI. Because most jobs still require meaningful human involvement, it concludes that transformation is considerably more likely than outright redundancy. That distinction matters, particularly for organisations deciding what their future technology workforce should look like.
Jobs are collections of different activities. A software engineer might write code, investigate defects, review architecture, speak to users, mentor colleagues and make decisions about security or scalability. AI may be able to generate some of the code. That does not mean it can assume responsibility for the entire role.
The work most vulnerable to automation tends to share several characteristics: it is repetitive, rules-based, digitally delivered and relatively easy to assess against a defined correct answer. Within technology teams, that could include:
These activities will not necessarily disappear completely, but organisations may need fewer people performing them manually. The value of the role then moves elsewhere: towards reviewing outputs, managing exceptions, understanding context and accepting responsibility for decisions.
Software engineering is usually one of the first professions mentioned in conversations about AI replacement, and there is some logic to the concern. Coding assistants can already generate functions, documentation, tests and suggested fixes. The narrow role of translating detailed specifications into standard code is under genuine pressure.
Writing code, though, is only part of software development. Engineers must still understand user requirements, work within complex legacy environments, make architectural trade-offs, protect systems from security threats and determine whether AI-generated code is accurate, maintainable and appropriate.
Labour-market projections illustrate this distinction. The US Bureau of Labor Statistics expects employment for narrowly defined computer programmers to decline by 6% between 2024 and 2034, noting that repetitive programming work is increasingly being automated. At the same time, employment across software development, quality assurance and testing is projected to grow by 15%, with AI itself contributing to demand for new software.
The likely outcome, then, is not the end of software engineering. It is the continued decline of coding as an isolated production activity. The engineers who remain valuable will be those who can use AI to move faster while applying the systems thinking, technical judgement and commercial understanding that the technology cannot provide independently.
IT support becomes service orchestration
AI can resolve common queries, search knowledge bases and guide users through standard troubleshooting processes. That will reduce the volume of basic work reaching first-line support teams, but it also increases the importance of people who can manage escalations, investigate unusual incidents and improve the systems behind the service.
The traditional service-desk analyst may evolve into an automation-enabled support specialist, service operations analyst or AI knowledge manager.
Manual QA becomes quality engineering
AI can generate test cases, execute repetitive tests and identify common patterns in defects. Quality, however, is not simply a matter of running more test scripts. Organisations still need people who can identify product risks, design testing strategies, explore unexpected behaviours and decide whether a release is genuinely ready.
Demand is likely to shift away from purely manual execution and towards test automation, exploratory testing and broader quality engineering.
Analysts become decision partners
AI is increasingly capable of producing dashboards, querying datasets and generating first-pass commentary, which reduces the value of simply producing reports. The analyst of the future will be expected to frame the right question, challenge the data, recognise weak assumptions and translate findings into decisions. Technical fluency will remain important, but it will increasingly be combined with commercial understanding and stakeholder influence.
Project coordination becomes delivery leadership
Meeting notes, progress reports, action lists and schedule updates are all highly automatable. Judgement is not. Programme and project professionals will spend less time maintaining administrative artefacts and more time resolving dependencies, managing risk, challenging priorities and creating alignment across teams. The role does not disappear; its centre of gravity moves from coordination to leadership.
Cybersecurity becomes more adversarial
AI can summarise logs, classify alerts and recommend responses to recognised threats. Attackers can use the same technology to operate faster and at greater scale. Cybersecurity professionals will therefore need to move beyond routine alert handling, with threat modelling, incident leadership, adversarial thinking and risk-based decision-making becoming even more valuable. AI will be part of the cyber team, but it will also be part of the threat landscape.
Technology has repeatedly removed certain tasks while creating new areas of specialisation, and AI appears to be following the same pattern. Organisations are beginning to recruit for roles such as:
These are not purely technical positions. They sit between technology, operations, risk, product and organisational change, and their purpose is to determine where AI should be used, how people should interact with it and what controls are required.
PwC's 2026 research found that technology, media and telecommunications had the highest AI hiring intensity of any sector, with nearly one in eight new roles related to AI. It also found that the skills required in the most AI-exposed jobs are changing more than twice as quickly as those in less-exposed roles. The opportunity is not limited to hiring machine-learning specialists; it is hiring people who can connect AI capability to genuine organisational outcomes.
AI's impact on junior roles deserves particular attention. Early-career technology professionals have traditionally developed by completing relatively straightforward tasks: fixing small defects, producing reports, manually testing applications or handling common support queries. Many of those are precisely the activities AI can perform most easily.
This creates a potential talent-pipeline problem. If organisations remove too much junior work, where will their future senior engineers, analysts and technology leaders come from? The answer should not be to stop recruiting entry-level talent. It should be to redesign how that talent develops.
Junior employees will need earlier exposure to AI tools, technical review, problem-solving and stakeholder communication. Employers will need stronger mentoring, practical rotations and structured opportunities for people to learn how to evaluate AI-generated work, not merely produce work that AI can already handle. The career ladder may become shorter and steeper. Organisations still need to give people a way to start climbing it.
The wrong response is an indiscriminate hiring freeze on the assumption that AI will eventually do everything. The more effective approach is to examine roles at task level. For each position, organisations should determine:
The World Economic Forum expects significant labour-market disruption by 2030, but projects that 170 million jobs will be created while 92 million are displaced, a net increase of 78 million. It also reports that 77% of employers plan to up skill their workforce in response to AI, while almost half expect to redeploy people from exposed roles into other parts of their organisation.
The organisations best positioned for this transition will not simply cut roles. They will redesign work, retrain existing people and recruit the capabilities their future operating model requires.
Will AI replace software developers?
No, not in the near term. AI can generate code, but software development also involves understanding requirements, managing legacy systems, making architectural decisions and judging whether AI-generated code is fit for purpose. The US Bureau of Labor Statistics projects a 6% decline in narrowly defined programming roles by 2034, alongside 15% growth in broader software development and quality assurance roles.
Which technology jobs are most at risk from AI?
Roles built around repetitive, rules-based tasks face the most disruption, including first-line IT support, manual regression testing, basic reporting and routine documentation. These tasks are likely to be automated or accelerated, but the roles themselves tend to evolve rather than disappear, shifting towards oversight, exception-handling and strategic input.
How should organisations prepare their technology teams for AI?
Rather than freezing hiring, organisations should assess roles at task level, identifying what AI can automate, what it can accelerate, what must remain human-led and what new capabilities are needed. The World Economic Forum found that 77% of employers plan to up skill their workforce in response to AI, while nearly half expect to redeploy staff from exposed roles.
AI will reduce demand for some types of routine work. It may enable smaller teams to deliver more, compress certain career pathways and increase expectations at every level of seniority. But organisations will still need people who can build systems, question outputs, protect customers, manage risk and turn technology into business value.
The question for employers is therefore not "how many people can AI replace?" It is "what combination of people, skills and technology will help us perform better?"
Answering that requires more than filling vacancies against an existing organisation chart. It requires a clear view of how roles are changing, which capabilities will become scarce, and what the future team needs to deliver.
TRIA helps organisations find the right technology talent through proven recruitment solutions, from key individual hires to complete teams. In an AI-enabled economy, competitive advantage will not come from choosing between people and technology. It will come from bringing the right people and technology together.
As a founder of TRIA, Harriet is proud of the company's impact in transforming recruitment through strategic insight and deep market understanding. Her leadership style is characterised by a focus on sustainable growth and the development of long-term client relationships.
Connect on LinkedIn
Written by Harriet Kirkpatrick
Written by Harriet Kirkpatrick
Written by Sean Hanly
