Databricks 10月27日 16:22
2025年风险管理新趋势
index_new5.html
../../../zaker_core/zaker_tpl_static/wap/tpl_guoji1.html

 

2025年Gartner风险管理报告指出,风险性质正以比许多组织适应能力更快的速度变化。人工智能的兴起、监管需求的扩大以及日益分散的数据环境意味着风险领导者必须重新思考他们的韧性方法。报告强调了几个关键主题,包括数据碎片化导致的风险盲点、数据治理向一线能力的转变、人工智能治理在应对人工智能风险中的作用、人工智能安全人才短缺的问题,以及将敏捷性和弹性集成到人工智能风险管理策略中的必要性。领先企业通过建立统一的数据基础、实施统一治理、嵌入人工智能治理和利用自助式洞察工具来应对这些挑战。

📊 数据碎片化导致风险盲点:风险管理中的盲点通常源于缺乏可见性而非努力不足。风险和安全数据分散在业务部门、IT系统和供应链中,这种碎片化使得早期检测微弱信号或快速响应几乎不可能。领先企业如Zillow、GM Financial和Shell通过建立统一的数据基础来解决这个问题。

🔒 数据治理成为一线能力:随着人工智能、云扩展和监管审查的融合,治理正成为一个前线能力和业务推动者。统一数据治理为企业提供了可信的数据源,支持更快行动、更自信的决策,并能在不引入新风险的情况下快速响应。例如,IQVIA和Bradesco Bank通过Databricks实现了数据治理和业务敏捷性的提升。

🤖 人工智能治理应对人工智能风险:人工智能具有两面性,既能通过自动化和机器学习改变风险检测、监控和缓解方式,也带来了可解释性、合规性和模型输出治理的新风险。组织需要在人工智能工作流程中嵌入治理,持续评估模型的准确性和偏差,并确保基础数据的安全和可信。DraftKings和McDonald’s通过Databricks实现了人工智能驱动的风险管理和业务决策优化。

👥 人工智能安全人才短缺问题:随着攻击面的扩大和监管的增多,风险和安全团队面临做更多事却资源更少的挑战。自助式洞察、自动化调查和高保真信号可以使团队效率倍增,将分析师从低价值警报中解放出来,专注于高影响威胁。美国海军通过Databricks模型审查400亿美元的交易,节省了200万工作小时,使团队能够专注于更高价值的风险和合规任务。

🚀 敏捷性和弹性集成到人工智能风险管理策略:Gartner讨论了需要快速适应新条件并保持弹性的敏捷风险计划。统一所有数据源——云、系统和格式——可以帮助组织预见问题而不是反应。统一治理为风险团队提供了这种可见性,并允许他们在需要时快速调整。Michelin和Adobe通过Databricks的数据网格和安全数据湖屋,实现了敏捷风险管理和实时网络安全分析。

Risk management has always been about anticipating the unexpected. But we feel the latest 2025 Gartner® 2025 Risk Report makes it clear: the nature of risk is changing faster than many organizations can adapt. The rise of AI, expanding regulatory demands, and increasingly fragmented data landscapes mean that risk leaders must rethink how they approach resilience.

Several themes from Gartner’s research stood out to us that align with modern risk models. Here are the key takeaways from the report and our perspective on what actions organizations can take to strengthen their own strategies:

1. Risk Blind Spots Start with Fragmented Data

One of the points from Gartner is that risk blind spots rarely come from a lack of effort; they come from a lack of visibility. Risk and security data remain scattered across business units, IT systems, and supply chains. That fragmentation makes it nearly impossible to detect weak signals early or respond with speed.

This experience is something we hear from our customers, too. When data lives in silos, teams are forced into a reactive stance, chasing alerts and reconciling conflicting sources rather than building a holistic picture of risk. The lesson is clear: a unified data foundation is the prerequisite for any modern risk strategy.

Here are examples of how leading companies are modernizing their data foundation:

    Zillow mitigates operational risk through automated dashboard analysis and AI-driven insights—helping its teams streamline on-call support and mission-critical operations by eliminating fragmented, manual processes.GM Financial built a unified customer view with strong governance.Shell manages all its analytics and AI workloads on a single platform—demonstrating how eliminating silos creates a single source of truth for decision-making.

2. Data Governance Moves to the Frontline

Governance has historically been seen as a cost center and compliance exercise—important, but peripheral to day-to-day operations. Gartner discusses a major shift: governance is becoming a frontline capability and business enabler.

Why? Because AI, cloud expansion, and regulatory scrutiny are converging. Organizations need greater assurance around access controls, data lineage, and accountability. Without it, innovation stalls under the weight of uncertainty.

We believe that organizations that unify data governance create a trusted source of truth, enabling faster action, more confident decision-making, and the ability to move quickly without introducing new exposure.

How enterprises are innovating faster with unified governance:

    IQVIA improved query performance and governance in healthcare analytics with Databricks, strengthening compliance and operational effectiveness.Bradesco Bank increased data integrity and business agility by building its in-house customer data platform with Databricks tools.

3. AI Governance Helps Combat AI Risk

Gartner mentions the double-edged nature of AI. On the one hand, automation and machine learning are transforming how risks can be detected, monitored, and mitigated. On the other, AI itself introduces new risks: explainability, compliance, and the governance of model outputs.

This is a balancing act we see across industries. The answer isn’t to slow down AI adoption, but to put robust guardrails in place from the start. That means embedding governance directly into AI workflows, continuously evaluating models for accuracy and bias, and ensuring the underlying data is secure and trusted. AI should amplify human expertise, not create new vulnerabilities and roadblocks.

Organizations are already striking this balance:

    DraftKings powers its real-time fraud detection pipeline with Databricks streaming and ML, enabling rapid and accurate threat identification.McDonald’s employs Databricks machine learning to optimize restaurant site selection and support high-stakes business decisions.

4. The AI Security Talent Gap Won’t Close on Its Own

Another key finding from this report is the persistent shortage of skilled risk and security professionals. As the attack surface expands and regulations multiply, teams are asked to do more with less.

Data can serve as a force multiplier. Teams equipped with self-service insights, automation for routine investigations, and high-fidelity signals can operate with far greater efficiency. Instead of wading through thousands of low-value alerts, analysts can focus on high-impact threats.

The Navy’s story illustrates this well: by building a model on Databricks to review $40B of financial transactions, they saved over 200,000 work hours, freeing teams to focus on higher-value risk and compliance initiatives.

5. Integrating Agility and Resilience into AI Risk Strategies

Gartner discusses the need for agile risk programs that adapt quickly to new conditions while maintaining resilience.

We believe agility starts with the data itself. Organizations that unify all sources— clouds, systems, formats—gain visibility to anticipate issues rather than react to them. The foundation of an agile risk program starts with unified governance, which provides this visibility and allows risk teams to pivot quickly when needed.

Here’s how two leading companies are approaching modern risk management:

    Michelin demonstrates the importance of agile risk management with its adoption of a Data Mesh on Databricks, empowering business users and streamlining operations across ERP and analytics.Adobe leverages the Databricks security lakehouse to perform real-time, large-scale cybersecurity analysis—helping its teams adapt rapidly to new threats.

Modern Risk Management Is a Dynamic System

We feel the Gartner findings point to a fundamental transition. Risk management is about creating a dynamic system powered by unified data, governance, and responsible AI, and agility.

The winners will be those who:

    Treat governance as a core capability, not an afterthought.Break down data silos to eliminate blind spots.Harness AI responsibly to augment human expertise.Empower teams with tools that reduce fatigue and increase focus.Build for agility, so resilience becomes a competitive advantage.

Closing Thoughts

In our opinion, the Gartner report is a call to action for security and risk leaders everywhere. The risks we face—cyber, operational, financial, regulatory—are only becoming more interconnected. Meeting that challenge requires not just more controls, but smarter foundations: unified data, embedded governance, and AI that is both powerful and safe.

For a deeper look at the Gartner research and recommendations, we encourage you to read the full report.

Gartner Reports: Gartner, 2025 Gartner® 2025 Risk Report, Avivah Litan, Max Goss, Sumit Agarwal, Jeremy D'Hoinne, Andrew Bales, Bart Willemsen, 18 February 2025

Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.

Fish AI Reader

Fish AI Reader

AI辅助创作,多种专业模板,深度分析,高质量内容生成。从观点提取到深度思考,FishAI为您提供全方位的创作支持。新版本引入自定义参数,让您的创作更加个性化和精准。

FishAI

FishAI

鱼阅,AI 时代的下一个智能信息助手,助你摆脱信息焦虑

联系邮箱 441953276@qq.com

相关标签

风险管理 人工智能治理 数据治理 敏捷性 韧性
相关文章