The threats facing organisations as they seek to implement AI tooling into their business are growing on a daily basis. In this paper, we seek to provide an overview of the common attack tactics and detail the high-level testing methodology we are currently utilising to deliver pentests against AI to protect both iSTORM® and our clients from the security risks associated with the use of AI tooling.
According to the ONS Business and Insights Conditions Survey (BICS) approximately one in six UK businesses (16%) are currently implementing at least AI application.
In this paper, we review the key threats associated with use of AI, providing an overview what each attack tactic would look to achieve We then discuss a high-level testing approach which are currently utilising within iSTORM®.
Access or download a copy of the full paper here: iSTORM – AI & LLM pentesting whitepaper FINAL 17.07.24 v1.3
