AI systems automate at least 20 percent of all current workforce tasks globally before 2030.
Concerns over AI taking over drive frequent queries regarding workplace displacement. Resolution requires a reputable economic body like the ILO or OECD to verify that one-fifth of standard occupational tasks have been fully automated by AI.
Across 6 frontier models answering independently, the panel thinks this is unlikely: the median is 15%.
They broadly agree, 18 points between Gemini at 23% and Claude at 5%.
Nothing is settled until 2 Nov 2031, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GeminiGoogle
Enterprise adoption and physical labor barriers make full automation of 20% of all global tasks unlikely in just five years.
- Most global labor is in developing economies and physical sectors like agriculture, construction, and care with low AI exposure.
- OECD and ILO task-level analyses rarely confirm full automation of complete occupational tasks even when augmentation is high.
- Rapid deployment of autonomous multimodal agents across white-collar sectors before 2028 would be required to shift these odds.
23% -
GPTOpenAI
This hinges on a credible global measurement showing full task automation, not merely AI assistance or exposure.
- ILO/OECD would need harmonized occupation-task surveys across economies and a defensible definition of “fully automated” by 2031.
- Historical technology diffusion takes decades; OECD and ILO measures usually report adoption, exposure, or task transformation rather than completed automation.
- Odds rise sharply if major employers deploy autonomous agents at scale and the ILO publishes a task-level global accounting standard.
18% -
GrokxAI
OECD/ILO must certify 20%+ of global tasks fully AI-automated by end-2029.
- OECD 2023 data show ~25% task exposure in rich countries but realized automation remains single-digit; global physical/agricultural work lags further.
- Full automation (zero human) plus developing-world scale-up has no 3-year precedent; bodies also delay and hedge language.
- Mass humanoid-robot + frontier-model rollout 2028-29 would be the main upward surprise.
18% -
DeepSeekDeepSeek
Whether ILO/OECD can verify 20% of tasks fully automated by 2030, given adoption lags and measurement gaps.
- ILO/OECD task-level automation estimates are survey-based and lag 2-3 years, so a 2030 threshold would only be verifiable around 2032-2033.
- Adoption is concentrated in software, finance, and customer service; agriculture, construction, and care work—over half of global employment—show minimal task automation.
- The 2024 ILO found only ~5-10% of tasks in high-income countries were AI-exposed, and actual automation lags exposure by years.
12% -
QwenAlibaba
Requires ILO/OECD to certify full automation of 20% of tasks globally in ~3 years—a bar no adoption curve has cleared that fast.
- ILO and OECD measure actual deployment, not capability; their 2023-24 reports peg real task displacement in low single digits globally.
- Full automation (zero human involvement) is far stricter than augmentation; even McKinsey's aggressive 2023 estimates project <10% of tasks fully automated by 2030.
- Global scope drags the average down—developing-economy adoption lags frontier economies by 5-10 years, making a 20% worldwide figure nearly impossible by 2029.
9% -
ClaudeAnthropic
Needs a formal ILO/OECD verified metric showing 20% task automation globally—no such metric exists or is tracking toward that.
- No international body currently measures or has committed to measuring 'percent of tasks automated' at global scale with a 20% threshold.
- Existing estimates (OECD, McKinsey, Goldman) suggest partial task augmentation, not full automation, and vary widely with no consensus methodology.
- Global diffusion to developing economies, regulatory friction, and integration lags make full automation of one-fifth of tasks by 2031 implausible in this timeframe.
5%
6 of 6 models answered · 18 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 3 Nov 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.