Computers using artificial intelligene, rather than conmen, will be behind most instances of fraud in the future, according to research by KPMG which found criminals turning to robotics in an effort to avoid detection.

KPMG's report, Profile of a Fraudster is based on analysis of 596 fraudsters investigated by the firm between 2011 and 2013. Rather than an increase in the number of 'traditional fraudsters' (identified as 36 - 45 years of age, acting against their own organisation and in an executive positions), KPMG says future threats will come from so-called 'seeker bots' capable of self-learning and self-replicating using artificial intelligence.

These 'bots' will be designed to continuously test a company's cyber defences and on finding a gap will analyse the potential for fraud and then launch a highly specialised 'attack bot', uniquely designed to suit the type of business, size, infrastructure and data setup of the victim. The aim will be to remove assets to a virtual delivery location which can then be accessed by the fraudsters.

Hitesh Patel, UK head of forensic at KPMG, says: 'This is not science fiction, but a taste of things to come. We are already seeing highly trained hackers link up with the organised crime network and the faceless criminal is not far away. Cyber crime is already on the rise and we expect cyber-attacks and high-tech fraud to grow exponentially.'

KPMG's analysis suggests this may already be happening, with a rise in the number of fraud cases involving collusion, up from 32% in the 2007 survey, to 61% in 2011 and 70% in 2013. Patel said that the research also challenged the stereotype of the lone hacker motivated by ideology, since in many cases, perpetrators were highly respected (39% of all cases analysed), regarded as sociable (35%) and/or an extrovert (33%).