Over Two-Thirds Imagine An Exterior Overview of AI Implementations Is the Greatest Option to Determine AI Security and Safety Points
HackerOne, a pacesetter in human-powered safety, revealed knowledge that discovered 48% of safety professionals consider AI is probably the most important safety danger to their group. Forward of the launch of its annual Hacker-Powered Safety Report, HackerOne revealed early findings, which embody knowledge from a survey of 500 safety professionals. In the case of AI, respondents have been most involved with the leaking of coaching knowledge (35%), unauthorized utilization of AI inside their organizations (33%), and the hacking of AI fashions by outsiders (32%).
When requested about dealing with the challenges that AI security and safety points current, 68% mentioned that an exterior and unbiased evaluation of AI implementations is the simplest approach to determine AI security and safety points. AI pink teaming provides the sort of exterior evaluation by way of the worldwide safety researcher neighborhood, who assist to safeguard AI fashions from dangers, biases, malicious exploits, and dangerous outputs.
“Whereas we’re nonetheless reaching business consensus round AI safety and security finest practices, there are some clear techniques the place organizations have discovered success,” mentioned Michiel Prins, co-founder at HackerOne. “Anthropic, Adobe, Snap, and different main organizations all belief the worldwide safety researcher neighborhood to offer knowledgeable third-party perspective on their AI deployments.”
Additional analysis from a HackerOne-sponsored SANS Institute report explored the influence of AI on cybersecurity and located that over half (58%) of respondents predict AI might contribute to an “arms race” between the techniques and strategies utilized by safety groups and cybercriminals. The analysis additionally discovered optimism round the usage of AI for safety workforce productiveness, with 71% reporting satisfaction from implementing AI to automate tedious duties. Nevertheless, respondents believed AI productiveness beneficial properties have benefited adversaries and have been most involved with AI-powered phishing campaigns (79%) and automatic vulnerability exploitation (74%).
“Safety groups should discover the most effective functions for AI to maintain up with adversaries whereas additionally contemplating its present limitations — or danger creating extra work for themselves,” mentioned Matt Bromiley, Analyst at The SANS Institute. “Our analysis suggests AI ought to be considered as an enabler, moderately than a risk to jobs. Automating routine duties empowers safety groups to give attention to extra strategic actions.”
Join the free insideAI Information newsletter.
Be part of us on Twitter: https://twitter.com/InsideBigData1
Be part of us on LinkedIn: https://www.linkedin.com/company/insideainews/
Be part of us on Fb: https://www.facebook.com/insideAINEWSNOW
