AI models will be able to complete 20-hour software engineering tasks in 2026
5
Ṁ1kṀ5892027
55%
chance
1H
6H
1D
1W
1M
ALL
This is a prediction from https://www.understandingai.org/p/17-predictions-for-ai-in-2026
It will resolve according to the judgement of the authors. If they don't publish a retrospective by the end of February, it will resolve according to my judgement.
This question is managed and resolved by Manifold.
Market context
Get 
1,000 to start trading!
1,000People are also trading
Related questions
[ACX 2026] Will an AI model reach a 3 hour time horizon with 80% reliability during 2026?
51% chance
Will models be able to do the work of an AI researcher/engineer before 2027?
25% chance
AI time horizon of 20 hours by January 1st 2027?
80% chance
Will an AI system capable of doing tasks that take humans eight hours as determined by METR.org, exist by 2027
91% chance
AI model training time decreases fourfold by mid-2027?
36% chance
AI for AI 1: By when will there be an AI that can do "intern-level" ML engineering?
AI for AI 2: By when will there be an AI that can do "senior-level" ML engineering?
AI "devops" #1: Will there be an AI that can onboard as a dev by 2028?
82% chance
Benchmark Gap #9: Once a model solves current software engineering benchmarks, how long until humans don't code?
8.1
Will the first AI model that saturates Humanity's Last Exam be employable as a software engineer?
45% chance