Teams Transcript accessing user/company data
Lauri Kiviniemi
0
Reputation points
During a Teams meeting where non-english language is spoken and transcript is enabled, in the transcript we can find confidential and private email addresses that are not found using any search engine and are not participants of the meeting that was transcripted.
Also random names/surnames of current and ex-employees can be found at random intervals, again not connected to the actual conversation.
Seems like when transcript cannot translate correctly from non-english to english the "AI" tries to fill the gap with information from our company.
Anyone seen something similar?
Sign in to answer