Research finds crowdsourcing is vulnerable to malicious behavior

New research has found that malicious behaviour is the norm in crowdsourcing competitions — even when it is in everyone's interest to cooperate.

Crowdsourcing provides the ability to accomplish information-gathering tasks that require the involvement of a large number of people, often across wide-spread geographies, expertise, or interests.

However, researchers from the University of Southampton and the National Information and Communications Technology Australia (NICTA) found that a significant feature of crowdsourcing — its openness of entry — makes it vulnerable to malicious behaviour.

They observed such behaviour in a number of recent popular crowdsourcing competitions, through analysis based on the 'Prisoner's Dilemma' scenario, which shows why two purely 'rational' individuals might not cooperate, even if it appears that it is in their best interest.

Co-author Dr Victor Naroditskiy from the University of Southampton says: "Everyone from the 'crowd' can contribute to solving the task. This is exactly what makes crowdsourcing so powerful for solving tasks that are all but impossible for a closed group of individuals or an organisation.

"At the same time, the openness makes crowdsourcing solutions vulnerable to malicious behaviour of other interested parties. Malicious behaviour can take many forms, ranging from sabotaging problem progress to submitting misinformation. This comes to the front in crowdsourcing contests where a single winner takes the prize."

Surprisingly, making the attacks more expensive for the attacker is not an effective way of deterring them.

These findings, published in the Journal of the Royal Society Interface, will be important for the design of crowdsourcing competitions, as well as for firms that consider using a crowdsourcing to solve a task.