.
Interview. Before AI is commonplace in enterprise, corporate leaders must commit to a security testing regime that is tuned to the nuances and nuances of AI models. Chatterbox Labs CEO Danny Coleman, and CTO Stuart Battersby spoke at length to The Register about the reasons why companies are slow to move beyond AI pilot tests into production deployment. Coleman. “McKinsey is saying it’s a four trillion dollar market. How are you actually ever going to move that along if you keep releasing things that people don’t know are safe to use or they don’t even know not just the enterprise impact, but the societal impact?”
He added, “People in the enterprise, they’re not quite ready for that technology without it being governed and secure.”
In January, consulting firm McKinsey published a report examining the unrealized potential of artificial intelligence (AI) in the workplace.
According to the reportand “Superagency in the workplace: Empowering people to unlock AI’s full potential,” AI technology is gaining interest, but adoption has been slow.
You should not believe the hype of the model or guardrail vendors, as they will all tell you that it is safe and secure. The report states
“Leaders want to increase AI investments and accelerate development, but they wrestle with how to make AI safe in the workplace,” .
Coleman says that traditional cybersecurity and AI are colliding but most infosec teams still haven’t caught on, as they lack the background necessary to understand AI’s unique attacks surfaces. He cited Cisco’s purchase of Robust Intelligence, and Palo Alto Networks’ acquisition of Protect AI, as examples of players who have taken the right path.
Battersby stated that the key to deploying AI at scale for organizations is to adopt a regime of continual testing based on the actual performance of the AI service. He explained. Battersby says that this is crucial, as even authorized users can cause an AI system to do harmful things. Coleman. “And it’s not going to change anytime soon. It needs to be so much more layered.”
Battersby says that while this may incur some costs, constant testing can help to bring down costs – by, for example, showing that smaller, less expensive models are just as safe in certain use cases.
You can read the complete interview here…
