Inside the Briefcase
AI Adoption Hindered by Data Security Fears, But Caution Can’t Yield to Complacency
February 7, 2024 No CommentsSOURCE: Paro
Artificial intelligence (AI) and machine learning (ML) are increasingly crucial to business operations. However, as a recent survey by Paro, an AI-powered marketplace that connects businesses with finance, accounting and AI experts, shows, cybersecurity and data privacy risks are still giving companies pause.
The survey polled 250 U.S. C-Suite executives on their finance transformation plans and needs for the near future. While 83% of executives recognize the transformational nature of AI/ML in business processes, 42% have not yet implemented these technologies. Cybersecurity and data security stand out as the primary concerns associated with AI, with 54% of survey participants vocalizing apprehension around both related to AI/ML adoption.
As finance teams increasingly depend on AI/ML for essential functions like predictive analytics and process automation, the threat of proprietary data being leaked or hacked heightens. Despite AI’s promise, the survey signals many leaders are opting for discretion over urgency in adoption.
Rigorous Governance and Risk Management Is Needed Before AI Integration
Beyond general cybersecurity fears, 22% of Paro’s survey respondents said AI wasn’t advanced enough, or where it needs to be for them to adopt it yet. Recent high-profile data incidents from AI providers validate these worries over functionality and technical maturity.
For example, in 2023, OpenAI’s ChatGPT data breach leaked private user data and chat histories. Data breaches such as these from AI tools can lead to a vast amount of data becoming vulnerable, impacting data confidentiality on a company and individual level
Further, the survey found 67% of executives currently use AI for predictive analytics and forecasting. Despite AI demonstrating efficacy in these areas, experts caution that AI’s opacity and reliance on training data make it susceptible to data manipulation. If input data is tainted or corrupted by bias, it can skew or lead to incorrect outputs.
“While AI and machine learning offer immense potential, companies need to approach implementation carefully and strategically,” said Nate Varel, General Manager of Innovation at Paro. “As our survey shows, data security remains a major barrier for many businesses looking to adopt these technologies. We advise setting up robust governance frameworks and risk mitigation strategies before deploying AI, not after.”
Assessments Are Crucial, But Adoption Delays Are Not An Option
The risks are real, but businesses can’t lag on AI adoption indefinitely. Delaying adoption will reduce companies’ competitive edge in an increasingly AI-driven economy.
With deliberate governance and risk mitigation, companies can ethically tap AI’s far-reaching potential. Both AI and data security are fast-evolving fields. Staying current and proactive will enable businesses to act responsibly with these emerging opportunities.
Sorry, the comment form is closed at this time.