Data Quality and Combating Bias Emerge as Top Hurdles for AI Innovation, According to New Report

Amid surging AI adoption and increased regulatory oversight, global business executives call for better AI governance to mitigate risk


NEW YORK, Oct. 17, 2024 (GLOBE NEWSWIRE) -- Prove AI, a first-of-its kind AI governance tool powered by blockchain technology, today released a new report, The Essential Role of Governance in Mitigating AI Risk. The report, which polled 600+ global executives, found that AI governance is seen as a critical area of investment for leaders to maintain responsible AI systems and mitigate risk. Among other highlights from the findings, executives see strong AI governance as a key step towards improving trust with end-users, and ensuring AI’s long-term ROI.

The report, commissioned by Prove AI and conducted by Zogby Analytics, surveyed CEOs, CIOs and CTOs of large companies across the US, UK and Germany. Of the organizations polled, the vast majority, (96%) are already using AI to support business operations, with the same tally (96%) reporting AI budgets will increase in the coming year. Top motivations for AI investment were increasing productivity (82%), operational efficiency (73%), stronger decision making (65%) and cost savings (60%); top AI use cases included customer service and support, predictive analytics, and marketing and ad optimization.

While investments continue to surge, business leaders are also clearly recognizing the additional risk exposure that AI brings to their organizations. According to respondents, the biggest deterrent to implementing new AI solutions are data integrity and security. This concern adds to existing AI performance issues that executives have already encountered, including:

  • Data quality issues (eg inconsistencies or inaccuracies): 41%
  • Bias detection and mitigation challenges in AI algorithms, leading to unfair or discriminatory outcomes: 37%
  • Difficulty in quantifying and measuring the return on investment (ROI) of AI initiatives, making it challenging to justify investments and prioritize projects: 28%

Executives exhibited confidence in the risk-management measures around AI applications, with 95% of respondents stating their organization is doing well with AI risk management today. However, in probing existing and upcoming priorities across risk management and AI governance, the report found:

  • Only 5% of executives say their organization has implemented any AI governance framework.
  • 82% of executives say that implementing AI governance solutions is a somewhat or extremely pressing priority; 85% cite plans to implement an AI governance solution by summer 2025.
  • Most (82%) participants are supportive of an AI governance executive order to provide stronger oversight, while 65% are somewhat or very concerned about IP infringement and data security.

As the data shows, this confidence wavers when executives factor in the pending enforcement of global regulations like the EU AI Act. De-risking AI is a top priority, and there is still work to be done. For these reasons, implementing and/or optimizing a dedicated AI governance strategy has emerged as a top priority.

“Executives are making themselves clear: AI’s long-term efficacy, including providing a meaningful return on the massive investments organizations are currently making, is contingent on their ability to develop and refine comprehensive AI governance strategies,” said Prove AI CEO, Mrinal Manohar. “The wave of AI-focused legislation going into effect around the world is only increasing the urgency; for the current wave of innovation to continue responsibly, we need to implement clearer guardrails to manage and monitor the data informing AI systems.”

Prove AI Accepting Early User Applications

Prove AI is accepting applications for its early access program through the end of Q4 2024; interested parties can apply on the Prove AI website. The solution will also be demonstrated at IBM TechXchange 2024 in Las Vegas, on October 22.

To access the full report, go to: https://casperlabs.io/report/governance-mitigating-ai-risk

About Prove AI
Prove AI is a comprehensive AI governance solution that enables certifiable and tamper-proof auditing for organizations building, training and running AI models. Led by a team of proven operators and technologists from companies including IBM, Amazon, Accenture, Bain Capital and Dropbox, Prove AI brings unprecedented visibility and trust to AI solutions so that organizations can embrace the technology with confidence and comply with all relevant regulatory requirements. Learn more at proveai.com.

Media Contact
LaunchSquad
proveai@launchsquad.com