Artificial Intelligence programmers have developed #MeTooBots to identify sexual harassment and digital bullying on emails and other conversations in companies. The bots are capable of monitoring and flagging harassing communication between colleagues. The bots were named after the #MeToo movement which was a movement against sexual harassments and assaults.
Makers said that it won’t be easy to teach them what sexual harassment looks like given its linguistic subtleties and grey lines. The bots use an algorithm trained to identify political bullying and sexual harassments in emails, documents and chats.
Then, the bots can analyse the text in the conversations for various indicators determining how likely it is to become a problem. Anything the AI reads as potentially problematic is sent to a lawyer or human resource managers for investigation.
The indicators which are deemed as red flags are the manufacturing company’s secret. A major concern posed by the bots is the confidentiality of data.
Prof Brian Subirana, a lecturer in AI at Harvard and MIT, said the idea of using AI to root out harassment is promising though the bots’ capabilities were limited. He added that if the software leaked any data by mistake, the internal conversation of a company would be out for the rival companies to access.
He further said that there’s a type of harassment that is very subtle and very hard to pick up. We have these training courses [about harassment] at Harvard, and it requires the type of understanding that AI is not yet capable of.
One of the manufacturer of the bots is a Chicago based AI firm called NexLP whose AI platform is used by more than 50 corporates including many law firms in London. Another AI startup, Spot, has created a chatbot that allows employees to anonymously report sexual harassment allegations. The bot also trained to give advice and ask sensitive questions to further investigate the case.