PROBLEM: Non-invasive brain-computer interfaces report what they think a person intended, but there is no standard check on whether the output matches the intent — so a wrong decode can move a cursor, send a message, or trigger an action the person never meant.
SOLUTION: An output verification standard: every non-invasive BCI output carries a confidence score, a confirmation step for irreversible actions, and a log of output versus confirmed intent, so errors are measured instead of assumed away.
EVIDENCE BASIS: DOCUMENTED — BrainNet (Jiang et al., Scientific Reports 9:6115, 2019) is the first multi-person non-invasive brain-to-brain interface: EEG records senders' decisions, TMS delivers them to a receiver, five groups of three averaged 81.25% accuracy — an accuracy figure, not certainty, which is why verification is needed.
| Idea priority | Low |
| Needed By | Not sure -- Just thought it was cool |
By clicking the "Post Comment" or "Submit Idea" button, you are agreeing to the IBM Ideas Portal Terms of Use.
Do not place IBM confidential, company confidential, or personal information into any field.