# Two-thirds of 100 senior AI leaders (HFS survey) say their business logic and tribal knowledge is fragmented or undocumented. Where does that live for your agents? — retained record

> Retained reddit record from reddit:AI_Agents.

Canonical: https://highsignal.app/data/records/98bf4d3802d88d18

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:AI_Agents)  published 2026-11-14 06:55 ingested 2026-11-15 02:34

[original source ↗](https://www.reddit.com/r/AI_Agents/comments/1xvvsra/twothirds_of_100_senior_ai_leaders_hfs_survey_say/)

## retained text



```
HFS Research published a survey on Nov 13 of 100 senior AI and data leaders. The findings that stood out:

* Two-thirds say business logic and tribal knowledge in their organisation is fragmented or undocumented
* 26% treat context as owned IP
* 38% have a funded initiative to address it
* 63% want workflow-level or systemic AI within 24 months, while HFS reports 59% of Fortune 1000 AI portfolios still sit at the task or use-case level
* HFS also says firms without what it calls an agent operating system hit a ceiling around five agents

Limits: 100 respondents, HFS is a research and consulting firm so these are its survey findings, and the five-agent ceiling is HFS's observation, not a universal rule. Nothing here shows missing context is the only reason programs stall.

The part worth discussing in this sub: an agent can follow a rule without knowing the reason behind it, and those reasons often sit in people's heads, old tickets and chat threads.

A 10-minute test on your own setup: pick one agent, list five decisions it makes, and write down the 'because' behind each one. Then check how many of those reasons exist somewhere the agent can actually read.

For people running agents on real workflows:

1. Where does that knowledge live for yours: prompts, a wiki, a vector store, a person?
2. Who keeps it current?
3. Did you hit a wall at a certain number of agents, and what was it?
```



document fetched 2026-11-15 00:18 · canonical https://reddit.com/r/AI_Agents/comments/1xvvsra/twothirds_of_100_senior_ai_leaders_hfs_survey_say
