Enhesa Research Finds Compliance Teams Skipping AI Verification Face Higher Fines
Compliance teams that fail to verify AI output are nearly three times as likely to face regulatory fines, according to Enhesa's research. This highlights the importance of expert review in AI-driven compliance decisions.
Original reporting · ishn.com
Original: “Compliance Teams That Skip AI Output Verification Fined Nearly Three Times as Often, New Research Finds”
Read “Compliance Teams That Skip AI Output Verification Fined Nearly Three Times as Often, New…” at ISHN →
How this was written. A Safety Kit wrote this article with a language model from ISHN’s reporting (model: @cf/meta/llama-3.3-70b-instruct-fp8-fast); it describes that reporting rather than adding to it. The original, linked above, is the full story.
Key points from ISHN’s report
- Compliance teams that skip AI verification are nearly three times as likely to face regulatory fines.
- Regulatory complexity is a major challenge for compliance teams, with 90% of teams reporting at least one business impact in the past year.
- AI is delivering speed and efficiency gains for compliance teams, but is less effective in anticipating future regulatory risks.
- The use of unverified AI output can have serious consequences, including incorrect filings and regulatory fines.
- Compliance teams should prioritize verification and review of AI output to avoid negative outcomes.
Compliance teams are increasingly relying on artificial intelligence to manage regulatory complexity, ISHN reported. However, many teams are acting on AI-generated guidance without proper review, which can lead to negative outcomes. Enhesa's research shows that 40% of compliance teams often or always act on AI output without expert review.
The research found that teams that skip verification are more likely to face regulatory fines. In fact, 37% of teams that do not verify AI output reported facing a fine, compared to 13% of teams that do verify. This discrepancy highlights the importance of expert review in ensuring compliance with regulatory requirements.
Regulatory complexity is a major challenge for compliance teams, with 90% of teams reporting at least one business impact in the past year. This includes increased legal costs, significant increases in workload, and delayed product launches. As a result, many companies are reconsidering or delaying entry into new markets.
Enhesa's research also found that AI is delivering speed and efficiency gains for compliance teams, but is less effective in anticipating future regulatory risks. While 48% of respondents reported improvements in speed and efficiency, only 26% reported improvements in anticipating future risks. This suggests that AI is primarily being used to react to regulatory changes rather than proactively manage them.
The use of unverified AI output can have serious consequences, including incorrect filings, regulatory fines, and failed audits. In fact, 71% of teams running fully autonomous AI reported a negative outcome, compared to 47% of other teams. This highlights the need for compliance teams to carefully review and verify AI output before acting on it.
Enhesa's CEO, Keith Berry, noted that AI can be a powerful tool for compliance teams, but only when used in conjunction with expert human counsel. He emphasized the importance of being able to defend the logic behind AI-driven compliance decisions, which is a challenge for many teams. In fact, only 32% of compliance leaders are very confident in their team's ability to defend these decisions.
The research suggests that compliance teams should prioritize verification and review of AI output to avoid negative outcomes. This includes implementing defensibility standards, such as source citation and audit trails, to ensure that AI-driven decisions can be trusted and defended. By taking a more cautious approach to AI, compliance teams can minimize the risk of regulatory fines and other negative outcomes.
Questions and answers
What is the risk of not verifying AI output in compliance decisions?
Compliance teams that skip verification are nearly three times as likely to face regulatory fines, according to Enhesa's research.
How are compliance teams using AI to manage regulatory complexity?
Many teams are using AI to react to regulatory changes, but are less effective in anticipating future risks. ISHN reported that 48% of respondents reported improvements in speed and efficiency.
What is the importance of defensibility standards in AI-driven compliance decisions?
Defensibility standards, such as source citation and audit trails, are critical in ensuring that AI-driven decisions can be trusted and defended, according to Enhesa's CEO, Keith Berry.
