AI tools automate more than 50% of routine IT support and software maintenance jobs before 2030.
Searchers frequently ask if AI will replace information technology roles. Resolves yes if recognized labor statistics show over half of baseline 2024 IT support tasks or roles are fully automated by AI.
Across 6 frontier models answering independently, the panel thinks this is unlikely: the median is 16%.
They do not agree: Gemini puts it at 38% and Claude at 4%, 34 points apart. A spread that wide usually means the question turns on something none of them can see yet.
Nothing is settled until 16 Oct 2031, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GeminiGoogle
Enterprise inertia and messy legacy systems delay full job elimination even as LLM agents absorb narrow Tier-1 ticket volume.
- BLS and Gartner show technology augments tasks far faster than it eliminates headcount, with tech jobs historically expanding post-automation.
- Fully replacing over 50% of roles requires autonomous action across bespoke internal stacks, which IT risk compliance actively limits.
- A breakout in reliable multi-modal autonomous agents resolving unstructured, on-prem hardware and access issues would push this over 60%.
38% -
GrokxAI
BLS-style stats must show >50% of 2024 IT-support/maintenance roles or core tasks fully AI-automated by end-2029.
- Help-desk and sysadmin employment grew through 2025 despite copilots; a 50% drop in three years has no postwar precedent.
- Vendors need end-to-end autonomous ticket-closers without human fallback, then firms must actually cut half the headcount—both lag capability by years.
- A 2027-28 BLS survey showing >40% contraction in SOC 15-1230/15-1240 would most change this view.
18% -
DeepSeekDeepSeek
Whether AI can eliminate over half of routine IT support tasks or roles by 2030, per labor data.
- Enterprise IT automation is incremental, not wholesale; most 2024 support roles persist through 2027 with tool-assisted, not eliminated, ticket triage.
- Labor stats lag 12-24 months, so demonstration of >50% automation must be observed by roughly mid-2030 to register by the resolution date.
- Past technology waves (RPA, cloud) cut specific tasks but not over half of IT support headcount within a decade, setting a low base rate this case would need to double.
18% -
GPTOpenAI
The bottleneck is proving full AI replacement in official task or role statistics, not deploying copilots widely.
- BLS and comparable agencies measure employment and occupations, not a standardized share of 2024 tasks fully automated by AI.
- IT support and maintenance remain human-accountable for access, outages, security, legacy systems, and escalation; past automation mostly changed staffing mix.
- A credible BLS/OECD task study documenting majority end-to-end automation before October 2031 would sharply raise this estimate.
14% -
QwenAlibaba
Official labor statistics showing >50% full automation of a job category within 5 years has essentially no historical precedent.
- BLS occupational data lags 1-2 years and has never recorded a >50% drop in a broad category attributable to a single technology within 5 years.
- Enterprise IT adoption cycles (procurement, compliance, legacy integration) typically take 3-7 years even for proven tools; full automation of edge-case-heavy support…
- A sharp reversal would require mass coordinated layoffs at Fortune 500s explicitly citing AI replacement of IT staff, which hasn't materialized at scale as of late 2026.
9% -
ClaudeAnthropic
Requires official labor statistics showing >50% task/role automation by 2030, which no agency currently measures or is likely to confirm.
- No recognized labor statistics body (BLS, etc.) tracks 'percent of IT tasks automated by AI' as a metric, so clean resolution is structurally unlikely.
- AI copilots are augmenting IT support/dev work heavily but full automation of majority of roles by 2030 exceeds realistic adoption curves seen in enterprise IT change…
- A sudden definitive labor study or government report explicitly quantifying this threshold would be the main thing that could flip this.
4%
6 of 6 models answered · 34 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 17 Oct 2026. Forecast by google/gemini-3.8-flash, anthropic/claude-sonnet-5, openai/gpt-5.6-terra, x-ai/grok-4.6, deepseek/deepseek-v4.1-flash, qwen/qwen3.8-max-0902 via OpenRouter.