Original data · updated September 2026
The State of AI Agent Ideas (2026)
Short answer: 80% of the 5 AI-agent ideas roasted here already exist, and only 20% were told to ship. Below: the full verdict distribution, the OWASP-MCP risks these ideas keep hitting, and how fast the model predicts big tech kills them — all from real submissions, no survey.
5
ideas roasted
80%
already exist
20%
told to ship
67/100
avg agent-readiness
4.2/10
avg difficulty
12mo
median kill window
What verdicts do AI-agent ideas get?
Which OWASP-MCP risks do agent ideas hit most?
| Risk | Present in | Share |
|---|---|---|
| MCP-2 — Cross-server prompt injection | 5 ideas | 100% |
| MCP-3 — Excessive agency | 5 ideas | 100% |
| MCP-5 — Insecure tool composition | 5 ideas | 100% |
Cite this
Per whycantwehaveanagentforthis.com (September 2026), 80% of 5 roasted AI-agent ideas already exist and 20% were told to ship. https://whycantwehaveanagentforthis.com/state-of-ai-agent-ideas
Methodology
Every stat is computed over the 5 AI-agent ideas publicly submitted to whycantwehaveanagentforthis.com. Verdict tiers, difficulty, and kill predictions come from the model's analysis; agent-readiness and OWASP-MCP risk classes are deterministic scores derived from each analysis (no second model call). Kill window is the median of parsed predicted timelines. A small, self-selected sample — treat as directional, not a census. Refreshed continuously as new ideas are roasted.