Dr Andrew Ng, The Artificial Intelligence Innovator and Pioneer Helping The Planet Embrace AI
Managing The Overlooked Risk: Emerging AI In Third-Party Vendor Applications
By Ilakiya Ulaganathan, Technology Risk and Control LeaderDriving Trusted Digital Innovation.
As AI and machine learning (ML) become standard features in many industries, a growing but often unnoticed risk is emerging. More and more third-party vendor applications are integrating AI capabilities, yet organizations frequently lack full visibility or control over how these technologies operate within their environments.
For those responsible for security and compliance, this shift isn't just another technical challenge; it represents a strategic risk. The real question isn't if vendors will use AI but, rather, how well organizations are prepared to manage and secure these hidden AI components.
The Challenge: Black-Box AI In Your Vendor EcosystemMost businesses today rely heavily on a web of SaaS products and external platforms to manage everything from customer relationships to HR and data analytics. Many of these vendors have started adding AI-driven features such as smart suggestions, predictive analytics, automated workflows and natural language processing.
However, the problem lies in the lack of transparency. Companies often don't know how these AI functions handle their sensitive data. Questions arise: Are enterprise datasets being used to train these models? Where is this data stored? Who actually owns the information derived from these AI processes?
Expanding Risks On Multiple FrontsThird-party AI introduces several risks, including:
• Data Exposure: Critical information could be inadvertently shared with external AI systems without proper safeguards or consent.
• Regulatory Compliance: Unnoticed AI usage might violate data protection laws such as GDPR or HIPAA or upcoming AI-specific regulations.
• Bias And Ethics: AI models trained on incomplete or skewed data can produce unfair or harmful outcomes, especially in sensitive areas like hiring or healthcare.
• Supply Chain Weakness: Security lapses or configuration errors at a vendor could expose your organization to breaches far beyond your direct control.
Updating Contracts To Reflect AI RealitiesOne crucial but often ignored layer of protection is contract management. As vendors add AI features, many existing agreements don't adequately address the new risks that come with AI and machine learning.
Organizations need to take a proactive stance by reviewing and updating vendor contracts to include specific AI-related provisions such as:
• Clear disclosures about AI usage and capabilities.
• Restrictions on using company data to train or improve AI models.
• Defined ownership of AI-generated insights and outputs.
• Requirements for transparency, explainability and audit rights.
• Explicit terms on data retention and deletion once contracts end.
Incorporating these clauses helps maintain control over your data and ensures accountability, even as vendors evolve their technology stacks.
Across industries, organizations are revisiting long-standing vendor contracts as providers quietly introduce GenAI features into existing platforms. Mid-cycle reviews are becoming more common to address AI-specific risks. Typical updates include requiring disclosure of AI/ML components and their decision-making role, placing restrictions on using client data for model training (even in aggregated or anonymized form), and clarifying ownership of AI-generated outputs, ensuring that insights derived from enterprise data remain the client's intellectual property.
To ensure oversight, a right to audit AI models for explainability and bias was introduced, along with data deletion guarantees enforceable at contract termination. These changes were not only about risk avoidance—they also enabled more transparent conversations with the vendor about responsible AI use.
Building AI Transparency Into Vendor ManagementManaging AI risk requires expanding traditional third-party risk management approaches. Consider these best practices:
1. Demand full disclosure of AI/ML capabilities. Ask vendors to openly share where AI is used, what data powers their models and whether they rely on proprietary or open-source algorithms.
2. Strengthen data handling clauses. Ensure contracts forbid unauthorized data use for AI training, mandate encryption in transit and storage and respect jurisdictional compliance.
3. Require explainability and auditing. Push for audit trails and insights into how AI systems make decisions, particularly for critical areas like finance or healthcare.
4. Implement ongoing monitoring. Use advanced tools to continuously observe vendor activity, flag anomalies and detect potential compliance gaps.
Governance Across Teams Is EssentialEffective AI governance requires collaboration across security, legal, procurement and business teams:
• Vendor Inventory: During onboarding, vendors should disclose any AI/ML usage. Flagged vendors go through a centralized review process. Inventories should be reviewed quarterly, using access logs and procurement data to ensure completeness.
• Risk Evaluation: Use an AI-specific risk rubric to assess vendors during onboarding and annual reviews. For high-risk cases, conduct deeper technical assessments and include contract clauses covering audit rights and data use restrictions.
• Stakeholder Education: Provide regular training to help teams ask the right questions about AI ethics, data handling and transparency—no technical background required.
Final Thoughts: Trust But VerifyWhile AI capabilities in third-party tools can drive innovation and efficiency, placing blind trust in vendors is no longer wise. Organizations must evolve their risk management frameworks to be AI-aware—demanding transparency, ethical use and contractual clarity.
The future of third-party risk management goes beyond simple service availability—it's about understanding and governing the AI-powered elements shaping your data and business decisions.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
95% Of Business Applications Of AI Have Failed. Here's Why
MirageC/Moment via Getty ImagesFollow ZDNET: Add us as a preferred source on Google.
ZDNET's key takeaways:Investment in generative AI may be booming, but most individual businesses using it have yet to see the payoff. In fact, a new MIT study found that 95% of enterprises attempting to harness the technology aren't seeing measurable results in revenue or growth.
Also: Gen AI disillusionment looms, according to Gartner's 2025 Hype Cycle report
The study, conducted by MIT's Networked Agents and Decentralized AI (NANDA) project, was based on interviews with over 150 business leaders and an analysis of 300 business deployments of generative AI.
"Just 5% of integrated AI pilots are extracting millions in value, while the vast majority remain stuck with no measurable P&L impact," the authors write in the report.
It paints a stark contrast between promises and reality: while tech developers are selling AI tools like agents as productivity boosters, NANDA's new report indicates that for all but a vanishingly small minority, the technology is having little to no effect on businesses' bottom lines. What accounts for the huge disparity?
What isn't working - and what couldIt largely boils down to a matter of bureaucratic inefficiency. Generative AI tools can provide efficiency gains in the hands of competent individuals, but when business leaders attempt to integrate them into existing, company-wide operations and workflows, they tend to throw a wrench into the organizational machinery.
Also: 71% of Americans fear that AI will put 'too many people out of work permanently'
The main reason for this, according to the report, is that the generative AI systems that most businesses are attempting to deploy internally and at scale lack the ability to seamlessly adapt with existing organizational workflows, ultimately making them more of a hindrance than an accelerant.
"The core barrier to scaling is not infrastructure, regulation, or talent. It is learning," the authors write. "Most GenAI systems do not retain feedback, adapt to context, or improve over time." While an ability to remember past interactions, customize outputs to different contexts, and learn over time are all key traits of AI, the authors are specifically referring to the context of the technology's use within enterprise-scale operations.
One of the implications of the new study therefore seems to be that in order for businesses to make the most of generative AI, they'd do well to take a bottom-up (allowing employees to experiment and discover their optimal mode of human-AI collaboration) as opposed to a top-down approach (forcing all employees to use a particular tool in a manner that's tightly controlled by executives and supervisors).
Also: Stop using AI for these 9 work tasks - here's why
Another trend that emerged from the study was flawed prioritization in the application of generative AI. Many businesses that were failing to profit from the technology were using it for marketing and sales, while the 5% that were using it successfully tended to do so through the automation of more fine-grained and mundane "back-office" tasks.
Based on their study, the authors predict that future success will belong to those businesses that deploy agentic and adaptable models in the right places, while those that choose a general, top-down approach will continue to be frustrated.
"The next wave of adoption will be won not by the flashiest models," they write, "but by the systems that learn and remember and/or by systems that are custom built for a specific process."
AI hype and cultural pressureOn its surface, the NANDA study seems to lend support to the belief that generative AI is nothing but a massive hype bubble that will soon pop, not unlike the short-lived corporate rush into the metaverse that preceded it. If such a massive proportion of businesses aren't seeing results, then surely that means the technology is being pedaled on empty promises, right?
Time will tell. For now, companies across the board are doubling down on their investments in AI, promising customers and investors that the rise of more agentic systems will usher in a golden age of prosperity, creativity, and leisure. At the same time -- and on the heels of a GPT-5 launch that received mixed reviews -- OpenAI CEO Sam Altman himself said he sees an AI bubble taking shape.
Also: 5 ways automation can speed up your daily workflow - and implementation is easy
Meanwhile, the widespread cultural embrace of AI means that companies are facing huge pressure to integrate the technology quickly -- or risk looking like dinosaurs. As NANDA's study indicates, this rush is, in many cases, apparently taking place at the expense of any kind of well-calculated plan, and as a result, investments in generative AI are leading many companies nowhere.
Even at the individual level, generative AI can be counterproductive in the long-term -- even while boosting productivity in the present. A recent study conducted by Workday, for example, found a correlation between heavy use of AI at work and employee burnout, while other studies find evidence that AI use degrades critical thinking skills.
Artificial Intelligence Stop using AI for these 9 work tasks - here's why How to use GPT-5 in VS Code with GitHub Copilot This is the fastest local AI I've tried, and it's not even close - how to get it Is ChatGPT Plus still worth $20 when the free version offers so much - including GPT-5?Generative AI In Supply Chain Management: From Theory To Operational Impact
Dileep Kumar Rai is a Global Supply Chain Optimization Expert, Oracle Fusion Cloud architect, and demand forecasting leader.
Supply chain management has evolved from basic transportation functions into a high-speed, data-driven environment that necessitates rapid decision making. As a professional who links digital transformation with supply chain innovation, I have personally witnessed companies adopt new technologies to address volatility, labor shortages and information management challenges. Generative AI has a subtle yet significant impact on supply chains.
Supply chain management (SCM) organizations view generative AI as a strategic advantage, whereas most people associate large language models (LLMs) with content creation. Supply chain professionals are now focusing on how artificial intelligence can provide immediate decision-making support for their operations. The adoption of cloud-based SCM systems demonstrates my experience with supply chain transformations, as I believe AI can effectively assist in real-time decision making.
The Standard View Of AI's Role In Supply Chains Requires ReconsiderationSupply chain professionals currently rely on dashboards, along with enterprise resource planning (ERP) systems and static reports, to manage their inventory management, procurement and logistics activities. The current data distribution pattern between multiple systems and departments creates both delays and inefficiencies, as well as decision fatigue, because teams must switch between them.
Generative AI functions as a digital co-pilot, simplifying complex data and converting it into natural language while adapting to changing conditions, rather than replacing human judgment. Technological development has surpassed theoretical boundaries. The technology now functions as a practical solution that meets people's daily operational needs.
Conversational ERP And SCM Applications Transform Use Cases Into Real-World Business ImpactGenerative AI implementations in real-world settings provide the following capabilities to users:
• Simple inventory information is presented in easy-to-understand alerts.
• People can use conversational dialogue to access procurement and planning system capabilities.
• The system generates automated vendor communication messages.
• AI guidance helps new SCM professionals complete their onboarding process more efficiently.
The implemented solutions decrease the workload associated with report interpretation, minimize supply chain disruption response times and enhance distributed team collaboration.
Akash Kadam, a mechanical engineer and expert in supply chain and manufacturing, conducted research to demonstrate how the integration of ChatGPT with ERP systems (download required) enables the creation of a conversational interface, according to his recent findings.
Research by Kadam demonstrated that his system utilized natural language to provide context-specific answers and initiate subsequent operations. His research illustrates how the model leverages existing infrastructure through API connections and prompt engineering, eliminating the need for sophisticated custom models. The system is well suited for mid-sized businesses, as it combines scalability with user-friendly features.
Why This Matters NowGenerative AI delivers practical value to supply chain operations due to its useful applications, rather than its innovative features. The top solutions from today operate through APIs on familiar platforms that support human expertise augmentation, rather than replacement.
Industry 5.0 will be achieved through human-machine collaboration, as outlined in the broader vision, which aligns with the current direction of the solution. AI systems require explainable algorithms with adaptive capabilities that seamlessly integrate into current workflow operations.
The system utilizes generative AI to enable users to interact with systems through human-like interfaces, thereby reducing the need for outdated legacy interfaces.
Emerging Trends To WatchGenerative AI continues to grow in popularity for three distinct areas, including:
• Compliance documentation and audit prep.
• The supply chain disruption analysis utilizes what-if scenario modeling as a planning tool.
• AI systems generate summary reports that facilitate collaboration among departments during their meetings.
Each application utilizes speed, context and accessibility as its fundamental connections. The analysis of data extends to include team support for interpreting its meaning and selecting future directions.
Agentic AI And The Next Evolution In SCMAgentic AI represents an exciting path that the future holds for organizations. These systems differ from traditional AI models in that they operate autonomously to set objectives and perform actions while adapting to supply chain changes without the need for constant human oversight. In supply chain management environments, AI agents can automate stock reorders through demand signal analysis, re-route shipments during disruptions and handle entire procurement processes. Supply chain leaders who grant AI decision-making authority will shift their management style from reactive to proactive.
Bridging Innovation And ExecutionThe primary challenge in adopting technology lies in the disparity between innovative ideas and their practical implementation. Groundbreaking research often remains confined to academic publications and experimental phases. Frameworks such as Kadam's serve organizations because they provide functional methods that enable testing and expansion across various implementations.
Supply chain leaders who need to handle ongoing change should focus on embedding intelligence into current tools rather than striving for ideal AI systems.
Final ThoughtsThe use of supply chain generative AI goes beyond just automation features. It helps teams speed up operations, make better decisions and reduce complexity, all while maintaining human connection in their work. Organizations that succeed the most will be those that see AI as an open system that functions clearly and reliably.
The first step toward starting your generative AI journey should be to build on your team's existing knowledge before improving accessibility, actionability and adding intelligence.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Comments
Post a Comment