My main concern about making data open would be if people misinterpret your data, which could then lead to invalid conclusions to be made in further study. These false conclusions could then come back to the original author and seen as their mistake, when it fact it's all due to a misinterpretation or miscommunication.

The recommendations to avoid or deal with this potential problem is to document how it should be interpreted and to be prepared to help and correct people if they have misinterpreted your data. Remain proactive in observing how your data is used, so that you can quickly refute the wrong interpretation.