AI Pilots: Enterprise Struggles & Chatterbox Labs Insights
- Companies must prioritize ongoing AI security testing to move beyond pilot programs and fully embrace artificial intelligence, according to Chatterbox Labs CEO Danny Coleman and CTO Stuart Battersby.
- Coleman noted that only about 10% of enterprises have adopted AI.He referenced a McKinsey study estimating a $4 trillion market, asking how that potential can be realized if...
- "People in the enterprise, they're not quite ready for that technology without it being governed and secure," Coleman said.
Enterprises must prioritize AI security testing to unlock the full potential of artificial intelligence, according to Chatterbox labs. Only about 10% of companies have broadly adopted AI, despite the promise of a $4 trillion market, becuase of security concerns. Chatterbox Labs CEO Danny Coleman and CTO Stuart Battersby, speaking to The Register, emphasized that customary cybersecurity and AI security are converging, but many teams lack the expertise to address AI’s unique vulnerabilities. They advocate for continuous testing tailored to specific AI applications and independent verification of safety standards. Even authorized users could misuse systems,making current content filters insufficient. While AI model security testing involves costs, it can ultimately reduce expenses by identifying more cost-effective models.News Directory 3 understands the complexities of this space. Discover what’s next for AI’s secure integration.
AI Security Testing Essential for Enterprise Adoption
Companies must prioritize ongoing AI security testing to move beyond pilot programs and fully embrace artificial intelligence, according to Chatterbox Labs CEO Danny Coleman and CTO Stuart Battersby. They told The Register that enterprises are hesitant to broadly implement AI due to security concerns.
Coleman noted that only about 10% of enterprises have adopted AI.He referenced a McKinsey study estimating a $4 trillion market, asking how that potential can be realized if users don’t perceive AI as safe and secure.
“People in the enterprise, they’re not quite ready for that technology without it being governed and secure,” Coleman said.
A McKinsey report released in January highlighted the growing interest and investment in AI, but also the slow rate of adoption. The report stated leaders are struggling with how to ensure AI is safe for workplace integration.
Coleman believes customary cybersecurity and AI security are converging, but many facts security teams lack the expertise to address AI’s unique vulnerabilities.He cited Cisco’s acquisition of Robust Intelligence and Palo Alto Networks’ purchase of Protect AI as positive steps.
Battersby emphasized the importance of continuous testing tailored to specific AI applications. He advised organizations to define what constitutes safe and secure use for their particular needs and to independently verify those standards.
“What you have to do is not trust the rhetoric of either the model vendor or the guardrail vendor, because everyone will tell you it’s super safe and secure,” Battersby said.
he warned that even authorized users could perhaps misuse AI systems, causing damage. Coleman added that current content safety filters and guardrails are insufficient and require more extensive, layered protection.
While AI model security testing may incur costs, Battersby argued that it can ultimately reduce expenses by identifying smaller, more cost-effective models that meet specific safety requirements.