All posts by Aslihan Buner, Kateryna Bozhenko

In today’s rapidly evolving AI landscape, strong governance is more critical than ever as organizations strive to harness the power of AI. Drawing on DataRobot’s ten years of experience in enterprise AI, we have dedicated ourselves to building an AI platform with the highest ranking for Governance Use Case among all 18 recognized vendors by Gartner^® in the market and a governance framework that we believe exceeds industry standards.

At DataRobot, we’ve always prioritized establishing a solid AI governance framework that ensures our customers can build, deploy and monitor generative and predictive AI assets with confidence. This framework helps teams maintain the quality and integrity of assets in production which is key in ensuring sustainable value. 

We believe this commitment has led us to be ranked the highest in Governance Use Case by Gartner^®, with an impressive 4.10 out of 5 governance score. In our opinion, this recognition is a testament to our unwavering dedication to upholding the highest standards of integrity, quality, and transparency across all AI operations.

The Growing Need for Strong AI and Data Governance

With the advent of generative AI, the demand for reliable governance has never been stronger or more urgent. As AI continues to become more deeply embedded across all sectors, the potential risks associated with its deployment grow accordingly. 

In 2023 alone, the AI industry saw a 40% increase in reported incidents related to data breaches and model bias, highlighting the urgent need for robust governance frameworks. According to a recent survey by PwC, 85% of AI leaders cite governance as their top concern, emphasizing the importance of trust, confidence, and the security of valuable intellectual property. 

At DataRobot, our AI governance capabilities are specifically designed to address these critical needs. Our platform provides comprehensive tools and protocols to bridge the confidence gap for our customers.

DataRobot enables the rapid and secure deployment of machine learning and generative AI applications into production within an average timeframe of 2 to 4 weeks. This accelerated deployment is facilitated by features such as automated compliance documentation, real-time risk management,  full model transparency, and most importantly strong guards and intervention methods.

This dual focus on governance and speed means that our customers can maintain a competitive advantage in AI without worrying about reputation damage or costly compliance issues. 

Key Governance Features that Set DataRobot Apart

AI has always been a team sport, and generative AI has made AI assets more accessible to a broader set of users, increasing the need for collaboration. Meeting high governance standards across all phases of the AI lifecycle — building, testing, production and management — is a must.

DataRobot Governance Umbrella

The DataRobot Governance Umbrella encapsulates our comprehensive approach to governance standards for both ML and GenAI development and management.

Our AI governance framework is designed to ensure that AI solutions are effective, efficient, and compliant, ultimately ensuring value with AI. It also extends compliance capabilities by ensuring that risk is mitigated across all AI assets throughout the end-to-end AI process:

• Build phase: Data scientists and AI practitioners lay the groundwork for creating robust AI solutions.
• Testing phase: Models undergo rigorous testing to ensure they meet our standards and perform reliably under various conditions.
• Production phase: Models are deployed and managed in a live environment.
• Monitoring and management phase: Oversight and governance tools help teams maintain the compliance, integrity and accuracy of AI solutions for operational excellence.

Our framework safeguards AI models and aligns them with operational and compliance goals. To do so, the DataRobot platform offers six unique features that empower each phase in the framework and make DataRobot stand out:

1. Visibility and Traceability: Full traceability of data, model lineage and versioning ensures that every change is tracked and documented which makes applications safe and useful.
2. Audits and Compliance Documentation: Automated generation of compliance reports and audit trails to meet regulatory requirements and fulfill transparency.
3. Unique LLM Evaluation and Testing: Detect potential risks utilizing both synthetic and real datasets to evaluate your predictive and generative AI models and benchmark performance.
4. CI/CD Testing: The ability to run prototype testing and evaluate ML or generative solutions with quality metrics to rank RAG experiments.
5. Real-time Intervention, Moderation and Alerting: Continuous monitoring with instant notifications and intervention capabilities enabled by guard models and metrics to address issues as they arise.
6. AI Catalog: Easily register, track and version all AI assets, whether they were built on or off the DataRobot AI Platform, all through a secure centralized hub.

Secure Collaboration Across Teams

From the build to the management phase, we prioritize data privacy, security, and efficiency as our customers like to move fast.

The Workbench in DataRobot provides an integrated environment for developing AI use cases with features for automated artifact registration for code, prompts, experiments and more. This helps accelerate the creation and iteration of useful AI models without sacrificing safety, or limiting collaboration. 

The Registry in DataRobot enables AI practitioners to catalog, version, and govern all AI assets ensuring more control over models. Encryption at rest and the option to bring your own key (BYOK) features ensure that our clients’ information is protected at all times, reinforcing trust and reliability.

Flexibility and Adaptability

The DataRobot AI Platform is one of the most open platforms for AI. We give our users the ultimate control and choice when it comes to their generative AI initiatives. The platform supports custom-built models, third-party APIs, and open source LLMs to prevent vendor lock-in and technical debt, and protect sensitive data.  

This flexibility and full governance behind company firewalls ensure that our customers can adapt their AI initiatives to meet evolving business needs with trusted security. Our platform also offers built-in GPU support to accelerate model training and processing, enabling data scientists to handle complex computations promptly.

At DataRobot, we offer equal governance for both predictive and generative AI, ensuring comprehensive oversight and control across all AI models. Our governance framework provides robust tools and protocols, including full model transparency, real-time risk management, and automated compliance documentation. Whether deploying predictive models or generative AI applications, our platform ensures that all AI assets adhere to the highest standards of security, integrity, and accountability. This balanced approach allows our customers to confidently and efficiently manage all of their AI initiatives, knowing that both predictive and generative models are governed with the same level of rigor and precision.

Recognized by Gartner, Trusted by Leaders 

Our governance framework has garnered praise from industry analysts, underscoring the real-world value and reliability that our platform provides. In addition to this, DataRobot has also been ranked highest in Governance Use Case by Gartner.

As much as this recognition from Gartner means to us, the most impactful feedback is from our customer community. Their testimonials highlight how our robust governance functionalities have positively impacted their AI initiatives, ensuring safe, successful and confident deployments:

Tom Thomas, VP of Data Strategy, Analytics & Business Intelligence at FordDirect

“DataRobot is an indispensable partner helping us maintain our reputation both internally and externally by deploying, monitoring, and governing generative AI responsibly and effectively.”

Arvind Thinagarajan, VP, Data Science & Analytics at Gannett | USA Today Network

“With DataRobot, we’ve already automated multiple steps in the machine learning lifecycle for hundreds of our models. These are models mostly in the realm of predictive AI as of now. This allows us to create efficiencies and saves time for my team of data scientists with steps like data pre-processing, model building, governance of those models, and measuring the performance of those models. We believe that our partnership can extend to the generative AI realm as well.”

At DataRobot, we are committed to enabling our customers to achieve their goals with confidence and excellence. Our Governance, recognized by Gartner and lauded by our customers, in our opinion, underscores our dedication to providing a reliable, transparent, and accountable AI platform. We are proud to be the trusted choice for organizations seeking to leverage AI responsibly and effectively.

Book a product tour and explore how our AI governance and compliance capabilities help you achieve value rapidly and scale AI use cases effectively.

Gartner Critical Capabilities^TM for Data Science and Machine Learning Platforms, Machine Learning (ML) Engineering, Afraz Jaffri, Aura Popa, Peter Krensky, Jim Hare, Tong Zhang, Maryam Hassanlou, Raghvender Bhati, Published June 24, 2024.

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally, and MAGIC QUADRANT and PEER INSIGHTS are registered trademarks of Gartner, Inc. and/or its affiliates and are used herein with permission. All rights reserved.

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.

This graphic was published by Gartner, Inc. as part of a larger research document and should be evaluated in the context of the entire document. The Gartner document is available upon request from DataRobot.

The post DataRobot Ranked #1 for Governance Use Case by Gartner®: Inside Our Trusted AI Governance Framework appeared first on DataRobot AI Platform.

In today’s data-driven world, ensuring the security and privacy of machine learning models is a must-have, as neglecting these aspects can result in hefty fines, data breaches, ransoms to hacker groups and a significant loss of reputation among customers and partners.  DataRobot offers robust solutions to protect against the top 10 risks identified by The Open Worldwide Application Security Project (OWASP), including security and privacy vulnerabilities. Whether you’re working with custom models, using the DataRobot playground, or both, this 7-step safeguarding guide will walk you through how to set up an effective moderation system for your organization.

Step 1: Access the Moderation Library

Begin by opening DataRobot’s Guard Library, where you can select various guards to safeguard your models. These guards can help prevent several issues, such as:

• Personal Identifiable Information (PII) leakage
• Prompt injection
• Harmful content
• Hallucinations (using Rouge-1 and Faithfulness)
• Discussion of competition
• Unauthorized topics

Step 2: Utilize Custom and Advanced Guardrails

DataRobot not only comes equipped with built-in guards but also provides the flexibility to use any custom model as a guard, including large language models (LLM), binary, regression, and multi-class models. This allows you to tailor the moderation system to your specific needs. Additionally, you can employ state-of-the-art ‘NVIDIA NeMo’ input and output self-checking rails to ensure that models stay on topic, avoid blocked words, and handle conversations in a predefined manner. Whether you choose the robust built-in options or decide to integrate your own custom solutions, DataRobot supports your efforts to maintain high standards of security and efficiency.

Step 3: Configure Your Guards

Setting Up Evaluation Deployment Guard

1. Choose the entity to apply it to (prompt or response).
2. Deploy global models  from the DataRobot Registry or use your own.
3. Set the moderation threshold to determine the strictness of the guard.

Configuring NeMo Guardrails

1. Provide your OpenAI key.
2. Use pre-uploaded files or customize them by adding blocked terms. Configure the system prompt to determine blocked or allowed topics, moderation criteria and more.

Step 4: Define Moderation Logic

Choose a moderation method:

• Report: Track and notify admins if the moderation criteria are not met.
• Block: Block the prompt or response if it fails to meet the criteria, displaying a custom message instead of the LLM response.

By default, the moderation operates as follows:

• First, prompts are evaluated using configured guards in parallel to reduce latency.
• If a prompt fails the evaluation by any “blocking” guard, it is not sent to the LLM, reducing costs and enhancing security.
• The prompts that passed the criteria are scored using LLM and then, responses are evaluated.
• If the response fails, users see a predefined, customer-created message instead of the raw LLM response.

Step 5: Test and Deploy

Before going live, thoroughly test the moderation logic. Once satisfied, register and deploy your model. You can then integrate it into various applications, such as a Q&A app, a custom app, or even a Slackbot, to see moderation in action.

Step 6: Monitor and Audit

Keep track of the moderation system’s performance with automatically generated custom metrics. These metrics provide insights into:

• The number of prompts and responses blocked by each guard.
• The latency of each moderation phase and guard.
• The average scores for each guard and phase, such as faithfulness and toxicity.

Additionally, all moderated activities are logged, allowing you to audit app activity and the effectiveness of the moderation system.

Step 7: Implement a Human Feedback Loop

In addition to automated monitoring and logging, establishing a human feedback loop is crucial for refining the effectiveness of your moderation system. This step involves regularly reviewing the outcomes of the moderation process and the decisions made by automated guards. By incorporating feedback from users and administrators, you can continuously improve model accuracy and responsiveness. This human-in-the-loop approach ensures that the moderation system adapts to new challenges and evolves in line with user expectations and changing standards, further enhancing the reliability and trustworthiness of your AI applications.

from datarobot.models.deployment import CustomMetric

custom_metric = CustomMetric.get(
    deployment_id="5c939e08962d741e34f609f0", custom_metric_id="65f17bdcd2d66683cdfc1113")

data = [{'value': 12, 'sample_size': 3, 'timestamp': '2024-03-15T18:00:00'},
        {'value': 11, 'sample_size': 5, 'timestamp': '2024-03-15T17:00:00'},
        {'value': 14, 'sample_size': 3, 'timestamp': '2024-03-15T16:00:00'}]

custom_metric.submit_values(data=data)

# data witch association IDs
data = [{'value': 15, 'sample_size': 2, 'timestamp': '2024-03-15T21:00:00', 'association_id': '65f44d04dbe192b552e752aa'},
        {'value': 13, 'sample_size': 6, 'timestamp': '2024-03-15T20:00:00', 'association_id': '65f44d04dbe192b552e753bb'},
        {'value': 17, 'sample_size': 2, 'timestamp': '2024-03-15T19:00:00', 'association_id': '65f44d04dbe192b552e754cc'}]

custom_metric.submit_values(data=data)

Final Takeaways

Safeguarding your models with DataRobot’s comprehensive moderation tools not only enhances security and privacy but also ensures your deployments operate smoothly and efficiently. By utilizing the advanced guards and customizability options offered, you can tailor your moderation system to meet specific needs and challenges. 

Monitoring tools and detailed audits further empower you to maintain control over your application’s performance and user interactions. Ultimately, by integrating these robust moderation strategies, you’re not just protecting your models—you’re also upholding trust and integrity in your machine learning solutions, paving the way for safer, more reliable AI applications.

The post How to Safeguard Your Models with DataRobot: A Comprehensive Guide appeared first on DataRobot AI Platform.