Developing a personalised roadmap for implementing best practices in AI governance

Colin Redbond, SVP Product Strategy, SS&C Blue Prism

Whether its customer chatbots or digital workers improving workflow, daily use of Artificial Intelligence (AI) and Intelligent Automation (IA) is under scrutiny.

With automation rising, nations are grappling with the ethical, legal, and societal implications and crafting AI governance laws, requiring business leaders to also prepare for these far-reaching changes.

The proposed EU’s AI act – considered the world’s first comprehensive law safeguarding the rights of users – is expected to regulate the ever-evolving needs of AI application developers in the EU and beyond.

Transparency and authenticity significantly influence brand perception, particularly among Gen Z – 32% of the global population. With high expectations, they only support brands aligned with their values.

Banks, auditors, and insurers, and their supply chains – adept at meeting legislation like Europe’s GDPR and Sarbanes Oxley in the U.S. – necessitate a similar approach with AI. Governance will influence everything from governments, robot manufacturing and back-office apps to healthcare teams using AI for data extraction.

Non-compliance could be substantial, with the EU suggesting fines of €30 million or six percent of global annual turnover, whichever is higher, so identifying AI integration points, workflows and risks is vital.

SS&C Blue Prism, through industry collaboration, offers guidance on governance roadmaps, ensuring businesses are well-prepared to meet evolving requirements while leveraging AI effectively.

Need for immediate action
The legislation also scrutinises automations, ensuring compliance as they innovate automated tasks, BPM data analysis, and business-driven automations. IA, with its auditable digital trail, becomes an ideal vehicle, providing transparent insights into actions and decisions, safeguarding record-keeping and documentation – crucial across the AI lifecycle.

Establishing and maintaining AI governance also fosters ethical and transparent practices with executives to employees, ensuring compliance, security, and alignment with organisational values, including:

• Top-down: Executive sponsorship ensures governance, data quality, security, and management, with accountability. An audit committee oversees data control, supported by a chief data officer.
  
• Bottom-up: Individual teams take responsibility for the data security, modelling, and tasks they manage to ensure standardisation, scalability enabling scalability.
  
• Modelling: Effective governance continuously monitors and updates performance to align with organisational goals, prioritising security in granting access.
  
• Transparency: Tracking AI performance ensures transparency and aids in risk management, involving stakeholders from across the business.

Frameworks for AI governance
Though standards are evolving, disregarding governance risks data leakage, fraud, and privacy law breaches so compliance and standardisation must be prioritised.

Governments, companies, and academia are collaborating to establish responsible guidelines and frameworks. There are several real-world examples of AI governance that – while they differ in approach and scope – address the implications of artificial intelligence. Extracts from a few notable ones are here:

The EU’s GDPR – not exclusively focused on AI – includes data protection and privacy provisions related to AI systems. Additionally, the Partnership on AI and Montreal Declaration for Responsible AI – developed at the International Joint Conference on Artificial Intelligence – focus on research, best practices, and open dialogue in AI development.

Tech firms like Google, Microsoft, IBM, and Amazon have created AI ethics guidelines, emphasising social good, harm avoidance, and fairness, while some countries have developed national AI strategies including governance.

Canada’s “Pan-Canadian AI Strategy” prioritises responsible AI development for societal benefit, focusing on ethics, transparency, and accountability. Establishing governance in your organisation involves processes, policies, and practices for AI’s responsible development, deployment, and use.

Reach governance greatness in 14 Steps
Government and companies using AI must incorporate risk and bias checks in mandatory system audits. Alongside data security and forecasting, organisations can adopt strategic approaches to establish AI governance.

• Development guidelines: Establish a regulatory framework and best practices for AI model development, including data sources, training, and evaluation techniques. Craft guidelines based on predictions, risks, and use cases.
  
• Data management: Ensure that the data used to train and fine-tune AI models is accurate and compliant with privacy and regulatory requirements.
  
• Bias mitigation: Incorporate ways to identify and address bias in AI models to ensure fair and equitable outcomes across different demographic groups.
  
• Transparency: Require AI models to provide explanations for their decisions, especially in highly regulated sectors such as healthcare, finance and legal systems.
  
• Model validation and testing: Conduct thorough validation and testing of AI models to ensure they perform as intended and meet quality benchmarks.
  
• Monitoring: Continuously monitor AI model performance metrics, updating to meet changing needs and safety regulations. Due to generative AI’s novelty, maintain human oversight to validate quality and performance.
  
• Version control: Keep track of the different versions of your AI models, along with their associated training data, configurations, and performance metrics so you can reproduce or scale them as needed.
  
• Risk management: Implement security practices to protect AI models from cybersecurity attacks, data breaches and other security risks.
  
• Documentation: Maintain documentation of the entire AI lifecycle, including data sources, testing, and training, hyperparameters and evaluation metrics.
  
• Training and Awareness: Provide training to employees about AI ethics, responsible AI practices, and the potential societal impacts of AI technologies. Raise awareness about the importance of AI governance across the organisation.
  
• Governance board: Establish a governance board or team overseeing AI model development, deployment and compliance with established guidelines that fit your goals. Crucially, involve all levels of the workforce — from leadership to employees working with AI — to ensure comprehensive and inclusive input.
  
• Regular auditing: Conduct audits to assess AI model performance, algorithm regulation compliance and ethical adherence.
  
• User feedback: Provide mechanisms for users and stakeholders to provide feedback on AI model behaviour and establish accountability measures in case of model errors or negative impacts.
  
• Continuous improvement: Incorporate lessons learned from deploying AI models into the governance process to continuously improve the development and deployment practices.

AI governance demands continuous commitment from leadership, alignment with organisational values, and adaptability to technological and societal changes. A well-planned governance strategy is essential for organisations using automation, ensuring compliance.

Establishing safety regulations and governance policies is vital to maintaining the security, accuracy, and compliance of your data. These steps can help ensure your organisation develops and deploys AI responsibly and ethically.