AI systems automate more than 50% of routine software engineering roles by 2029.
Advanced code generation models threaten software developer employment. Resolves yes if industry benchmark surveys show over half of entry-level coding positions are replaced or eliminated by AI by 2029.
Across 6 frontier models answering independently, the panel thinks this is unlikely: the median is 15%.
They broadly agree, 26 points between GPT at 31% and Claude at 5%.
Nothing is settled until 13 Oct 2029, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GPTOpenAI
It hinges on whether firms convert productivity gains into headcount cuts rather than cheaper, higher-volume software output.
- Employers must eliminate or cease backfilling over half of entry-level coding jobs, not merely deploy coding copilots, by October 2029.
- Past automation usually reshapes junior work before removing most roles; hiring cycles, legacy systems, and accountability slow displacement.
- A sharp drop in junior hiring through 2027–28, alongside credible cross-industry surveys attributing cuts to AI, would raise this materially.
31% -
GeminiGoogle
Enterprise adoption lags model capability; companies expand dev scope rather than eliminating headcounts outright.
- Jevons paradox historically drives more software creation when code costs fall, shifting roles rather than erasing headcount.
- Labor adjustments take 3-5 years to hit 50% benchmarks even with tools like Copilot, bounded by legacy systems and liability.
- A sustained net drop in entry-level hiring reports across major tech firms through 2027 would shift odds toward resolution.
27% -
GrokxAI
Entry-level coding headcount falls >50% only if firms stop hiring juniors en masse as agents become unsupervised-reliable.
- Firms and HR must cut junior reqs by 2028-29 hiring cycles; AI must first handle full tickets with <5% human fix rate.
- Historical tech displacement (ATMs, compilers, IDEs) never cut a role class 50% in 3 yrs; software demand grew instead.
- A 2027-28 agent that ships production PRs unsupervised would flip this; current copilots have not.
18% -
DeepSeekDeepSeek
This turns on whether firms actually eliminate entry-level headcount, not whether AI can do the work.
- Capability is already here: SWE-bench scores jumped from ~4% in 2023 to ~70%+ by 2025, so the binding constraint is adoption, not model skill.
- BLS projects software developer employment to grow ~17% through 2033, and 2024-2026 layoffs track interest rates and overhiring, not AI substitution.
- Firms like Salesforce and IBM publicly walked back AI-replacement claims in 2024-2025, showing the 'eliminate juniors' narrative outruns measured headcount data.
12% -
QwenAlibaba
Organizational inertia and the training-pipeline function of junior roles make 50% elimination in 3 years implausible.
- Even with rapid capability gains, enterprises need junior devs as a senior-dev pipeline; no major firm has publicly committed to eliminating that funnel.
- Historical base rate: no technology has eliminated 50%+ of a white-collar job category within 3 years of commercial availability (ATMs, spreadsheets, CAD all took 10+…
- Resolution requires industry benchmark surveys documenting the drop—surveys like Stack Overflow Developer Survey or BLS data lag reality and would need unambiguous…
9% -
ClaudeAnthropic
Needs a clear industry benchmark showing >50% of entry-level coding roles eliminated by AI within 3 years—unlikely by any measure.
- No standardized 'industry benchmark survey' currently tracks this metric precisely, making clean resolution unlikely even if displacement grows.
- Current evidence shows AI augmenting developers and slowing junior hiring growth, not eliminating over half of roles outright by 2029.
- Labor market data (BLS, LinkedIn hiring reports) would need to show a dramatic structural collapse in entry-level coding jobs specifically attributable to AI, which…
5%
6 of 6 models answered · 26 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 14 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.