Skip to content
Longterm Wiki

Agentic AI

Capabilities Researchactive

Research on AI systems that take autonomous actions: tool use, computer use, multi-step planning.

Wiki page →
Key Papers
18
Cluster: Capabilities Research

Tags

scope:field

Key Papers & Resources18

SWE-bench Official Leaderboards
AGILE Index on Global AI Safety Readiness
Gartner predicts 40%+ agentic AI projects will be cancelled by 2027
Precedence Research
2025 08 26 Gartner Predicts 40 Percent Of Enterprise Apps Will Feature Task Specific Ai Agents By 2026 Up From Less Than 5 Percent In 2025
First autonomous coding agent
10-15% of market by 2030
Gartner predicts
agentic AI market
Responsible Scaling Policy
SWE-bench Verified - OpenAI
Preparedness Framework
Frontier Safety Framework v2
Engineered prompts in emails
Shrestha Datta, Shahriar Kabir Nahin, Anshuman Chhabra, Prasant Mohapatra2025-10-27
Cooperative AI research
Sunil Arora, John Hastings2025-12-19
Misalignment or misuse? The AGI alignment tradeoff
Max Hellrigel-Holderbaum, Leonard Dung2025-06-04
FLI AI Safety Index Summer 2025
Anthropic: Recommended Directions for AI Safety Research