Managing AI risk with confidence

Is your organization using artificial intelligence?

Artificial intelligence is probably already being used somewhere in your organization. Whether it’s a chatbot or machine learning algorithm, AI programs need governance, discipline and ongoing care and maintenance to perform well, reduce risk and drive expected outcomes.

Ready to learn more?

Demystifying AI

loading-player

Playback of this video is not currently available

The three building blocks of AI? Data, algorithm and output.

Do you know:

  • Who owns the strategy for AI in your organization?
  • What team has governance and oversight over AI?
  • What controls are in place for AI?
  • What data is being used or created?
  • How data is being stored and protected?
  • What algorithms you're using from third party vendors—and how you're protected?

It’s critical to consider how AI affects your business from a financial, operational and customer perspective and how it impacts your reporting needs.

Managing AI with confidence

Here's what our risk and assurance services include:

Assess and recommend

Assessment and recommendation

Using the defined PwC Risk Framework, we provide management with a readiness assessment report, sharing observations and recommendations to enhance your AI program and prepare for future attestation reporting. We also design and can implement specific systems controls for all aspects of AI (e.g., process oversight, data, models, output, machine learning), where permissible.

icon

Algorithm performance

To help ensure performance meets expected goals we can perform AI testing. We review your models against documented policies and standards, and expected range of model outcomes. We also provide recommendations for tuning.

Continuous monitoring services

Continuous monitoring services

Our continuous monitoring services provide consistent oversight for risks in algorithms, minimizing disruptions and helping to mitigate risk, while also providing reporting for all stakeholders.

Governance and oversight

Governance and oversight

We can help build a robust AI governance and oversight program to give you a coordinated and disciplined approach to AI across your organization.

Bias review

Bias review

Through detailed testing of outputs against identified key risk domains, we can highlight potential bias in algorithm performance and provide recommendations for corrective actions.

Independent review

Independent review

Providing an independent, third-party perspective, PwC supports important new AI development, application and use, and can help to accelerate acceptance and adoption of critical new trusted uses of AI through third-party assurance services.

Audit attestation services

Audit attestation services

By incorporating AI processes into existing or new controls reports, management can obtain third-party assurance over design and operating effectiveness of controls that incorporate AI.

Contact us

Jennifer Lendler

Jennifer Lendler

Risk Assurance Managing Director – Emerging Technology & Innovation, PwC US

Chris  Thompson

Chris Thompson

Partner, PwC US

Follow us