Password resets and account admin
Self-service and automation own this already in modern organisations. It should not appear on any career plan.
Expected horizon: 0-2y
Risk90%
INFORMATION TECHNOLOGY · ISCO-08 3512
Will AI replace it support specialist jobs?
First-line support is the textbook AI use case: repeatable questions with documented answers, arriving as text. Chatbots, self-service portals and self-healing systems are absorbing that layer now, and ticket volumes reaching humans are falling in automated organisations. What holds is everything with a physical or human edge: hardware in hands, on-site presence, executives who want a person, incidents too tangled for scripts. The occupation is splitting cleanly. Scripted remote support shrinks hard. Field work, complex escalation and the administration of the support AI itself grow. Specialists should decide early which side of the split they are building toward.
How AI may change the work
A job is a mix of tasks. The title alone cannot show your personal risk.
First-line ticket triage
AI reads, categorises and answers routine tickets before humans see them. The queue that trained juniors is disappearing.
Expected horizon: 0-2y
Risk80%
Standard troubleshooting from known fixes
Known-error fixes execute automatically or guide users directly. Value moves to the problems without a documented answer.
Expected horizon: 0-2y
Risk70%
Hardware setup and physical repairs
Devices still need hands: builds, swaps, cabling, peripherals. Physical work is the role's most durable layer.
Expected horizon: 10y+
Risk25%
Escalating complex incidents
Knowing what is actually wrong, and who genuinely needs to act, separates professionals from ticket-routers. Build this skill deliberately.
Expected horizon: 2-5y
Risk45%
User training and hand-holding
Patient humans helping stressed users remain valued, especially around new tools. Ironically, AI rollouts create this work.
Expected horizon: 5-10y
Risk30%
Asset and licence management
Discovery and tracking automate steadily. Judgement on cost, renewal and standardisation decisions stays with people.
Expected horizon: 2-5y
Risk65%
On-site support for executives and events
Presence, discretion and calm under a board deadline. This white-glove work is relationship-based and firmly human.
Expected horizon: 10y+
Risk25%
What the score means
The headline score is high, but it belongs to a specific layer. Scripted, remote, text-based support is exactly the work generative AI absorbs first, and analyses of current capability place routine information-processing among the largest pools of automatable tasks [S7]. The physical and complex layers read differently: hands-on hardware, on-site presence and tangled incident work sit in the low-exposure categories of global research [S6]. There is also a compensating demand: organisations rolling out AI create waves of user confusion, tooling questions and system administration, which lands in support teams first. The occupation's honest summary is a shrinking level one, a durable field layer, and a new specialism running the support automation itself.
Your protection plan
Get physical or get deep, and start now. First, claim the hardware and field work. Device builds, site visits and executive support are the layers chatbots cannot reach, and they build the relationships that survive restructures. Second, convert triage experience into incident depth. Study the escalations, learn the systems behind the tickets, and become the person who solves what the script cannot. Third, administer the automation. Support chatbots, self-service portals and knowledge bases need owners who tune, test and improve them; managing the bot is the promotion path out of competing with it. Fourth, add a specialism with accountability: security operations, systems administration or identity management all reward exactly the judgement level one never taught.
Questions people ask
Is helpdesk work disappearing?
The scripted portion is, and quickly. Password resets, known fixes and routine triage now automate before reaching humans, matching global findings on routine information work [S7]. What remains and grows: physical support, complex incidents, user training during technology change, and administering the support automation itself.
What should a support specialist learn to stay relevant?
Choose depth over breadth of scripts. Systems administration, security operations or identity management add the judgement layer that survives. Alternatively, own the support AI: chatbots and portals need tuning, testing and content management, and the specialist who runs the bot outranks the ones it replaced.
Are on-site roles genuinely safer than remote helpdesk?
Yes, structurally. Physical presence, hardware handling and face-to-face support sit among the lowest-exposure task categories in global research [S6], while remote scripted support sits among the highest. If your current role is entirely remote and entirely scripted, treat that combination as the clearest relocation signal in IT.
Does AI adoption create any support work?
Substantial amounts, at least for now. Every rollout produces confused users, integration problems, access questions and new systems to administer. Support teams sit first in that firing line, which buys time and creates the bot-administration specialism. The work changes shape faster than it disappears.
Explore related roles
Software Developer
the neighbouring profession whose automation story rhymes with this one at a higher pay grade.
Data Analyst
a realistic destination for support specialists who grow into the data behind the systems.
Customer Service Representative
shares the scripted-interaction exposure without the technical escape routes.
Electrician
a reminder that physical technical trades hold the lowest exposure of all.
Executive Assistant
a parallel story of scripted tasks automating while trusted human service survives.
Sources and method
- [S7] MIT and Oak Ridge National Laboratory, Project Iceberg research report Accessed 2026-07-13.
- [S6] ILO Working Paper 140, Generative AI and Jobs: A Refined Global Index of Occupational Exposure Accessed 2026-07-13.
- [S1] UAE Government portal, UAE Strategy for Artificial Intelligence Accessed 2026-07-13.
- [S9] UAE Government portal, Artificial intelligence in government policies Accessed 2026-07-13.
- [S3] SDAIA, National Strategy for Data and AI Accessed 2026-07-13.
- [S4] Saudi Ministry of Human Resources and Social Development, Progress in the Saudi Labor Market Accessed 2026-07-13.
This page uses a reviewed task profile, not a generic job-title probability. Read the full methodology and limitations.