Sean Much, Co-CEO/CFO, AI & Partners, [email protected], +31(6)572 85579, Sean has an extensive background in the entertainment industry (e.g., film and art), and has a specialism in design. Alongside this, Sean has more than a decade of experience in the professional services sector, including holding the position of a tech accountant for 5 years. Sean knows about auditing and has helped with an IPO on the New York stock exchange. As well as being a compliance expert, he has deep expertise in implementation aspects of audit & assurance engagements and has been working with the largest global tech MNEs over the past 5 years.
Michael Charles Borrelli, Co-CEO/COO, AI & Partners, [email protected], +44(0)7535 994 132, Michael Charles Borrelli is a highly experienced financial services professional with over 10 years of experience. He has held executive positions in compliance, regulation, management consulting and operations for institutional financial services firms, consulted for FCA-regulated firms on strategic planning, regulatory compliance, and operational efficiency. In 2020, Michael set-up the operations model and infrastructure for a crypto asset exchange provider and has been actively engaged in the Web 3.0 and AI communities over the last 4 years. He currently advises a host of AI, Web3, DLT and FinTech companies.
Charles Kerrigan, Partner, Partner, CMS UK, [email protected], +44(0)7769 967516, Charles Kerrigan is a specialist in emerging technologies including crypto, digital assets, Web3 and AI. He works on corporate finance and venture capital transactions in crypto, tokenisation, NFTs, Web3 and DeFi. He works on consulting projects on blockchain and AI for public bodies, policymakers, standards institutions, and corporations.
The Blockchain Industry in the UK Landscape Overview names Charles as a “leading influencer in blockchain”. He is part of teams working on investing and setting standards for emtech in the UK, Europe, and the US. At CMS he is a Partner in a team that covers emtech. He has roles on the advisory boards of AI and crypto firms and of trade bodies. He is the Editor and co-author of Artificial Intelligence Law and Regulation (Edward Elgar, 2022). He is the Contributing Editor of AI, Machine Learning & Big Data (GLI, 2023). He was listed in The Lawyer Hot 100 2022 of “most daring, innovative and creative lawyers”.
Introduction
In recent years, the public sector has witnessed an unprecedented integration of Artificial Intelligence (“AI”) into its operations. From streamlining administrative tasks to enhancing citizen services, AI offers remarkable potential for innovation and efficiency. However, with great power comes great responsibility. As AI’s role in public organizations expands, so too does the need for effective governance and oversight. This article delves into the critical need for standardized contractual clauses when procuring high-risk AI solutions in the public sector, such as those proposed by the European Commission. The EU model contractual AI clauses, thoughtfully developed and peer-reviewed, offer a vital framework for public organizations procuring AI solutions. High-risk AI systems align with the proposed AI Act, while non-high-risk applications enhance AI trustworthiness. Flexibility exists for extending these standards to various algorithmic systems, recognizing accountability and transparency needs. These clauses don’t replace complete contracts but serve as a valuable addition. Future translations and an explanatory note will further aid their practical implementation.
It explores the rising significance of AI in public services and the compelling argument for standardized procurement clauses to safeguard responsible AI implementation.
The Significance of AI Procurement in Public Services
The deployment of AI in public organizations has been steadily growing, redefining the landscape of public services. AI is more than a technological innovation; it represents a strategic transformation in how governments and public institutions interact with citizens and address their needs. Public services can benefit immensely from AI, from optimizing administrative processes and resource allocation to improving service quality and accessibility. As governments increasingly turn to AI to enhance their services, the potential is vast.
However, this rapid adoption of high-risk AI solutions comes with inherent challenges and risks. Ensuring that these technologies are used responsibly and effectively is paramount. The need for standardized contractual clauses in AI procurement becomes evident as a crucial component in mitigating risks and ensuring accountability. These clauses provide a framework for public organizations to engage with AI vendors, stipulating the responsibilities and liabilities of each party, and establishing a foundation of trust and transparency. By doing so, public organizations can harness the potential of AI while safeguarding against potential pitfalls, ultimately delivering better services to their constituents.
Challenges and Risks in AI Procurement
The procurement of high-risk AI by public organizations is a venture laden with potential challenges and risks. As AI takes on an increasingly substantial role in public services, ethical concerns become more prominent, particularly in high-risk contexts. The implementation of AI in areas such as healthcare, law enforcement, or critical infrastructure demands meticulous consideration.
The challenges encompass a range of issues, including ensuring transparency in AI decision-making, protecting sensitive data, and mitigating biases. High-risk AI systems must be scrutinized carefully to avoid potential pitfalls. Ethical concerns emerge as AI systems often make decisions that have profound consequences for individuals and society as a whole.
Addressing these challenges requires comprehensive solutions. The proposal for standardized contractual clauses is a step towards creating a structured approach to AI procurement. These clauses provide a foundation for public organizations to navigate the complexities of AI procurement while safeguarding against risks and ethical concerns. They represent a proactive measure to ensure that AI is harnessed responsibly, ethically, and effectively in high-risk public service contexts.
The Role of Standard Contractual Clauses
Standard Contractual Clauses (“SCCs”) are familiar tools used across various domains to ensure compliance, ethical practices, and legal protection in contractual agreements. These standardized clauses are designed to streamline the negotiation process by providing pre-established terms that parties can adopt. Historically, they have been employed in fields like data protection to govern international data transfers, ensuring that data is handled in accordance with applicable regulations.
Their proposed introduction in AI procurement represents a significant step forward. In an era where AI is becoming increasingly critical to public organizations, SCCs can provide a structured framework for responsible AI implementation. These clauses offer a foundation for public entities to safeguard against the complexities and risks associated with high-risk AI procurement, fostering compliance, ethical practice, and accountability.
The Rationale Behind SCCs in AI Procurement
The introduction of SCCs in the procurement of high-risk AI by public organizations is grounded in a set of compelling ethical and legal imperatives. In an age where AI systems are becoming central to public services, ensuring ethical AI implementation has never been more crucial.
SCCs offer a structured approach to address critical ethical concerns, specifically those related to data protection, privacy, and transparency within AI systems. The integration of SCCs is necessitated by the inherent risks associated with AI, particularly in contexts where these systems can affect individuals’ lives and well-being. Without adequate safeguards, the potential for harm and adverse consequences, whether through biased decision-making or breaches of privacy, looms large.
By introducing SCCs in AI procurement, public organizations are proactively protecting the public interest. These standardized clauses function as guardrails, imposing ethical and legal boundaries that AI providers must adhere to, promoting responsible AI deployment, and safeguarding individuals and society as a whole.
Advantages of SCCs
SCCs play a pivotal role in the procurement of high-risk Artificial Intelligence AI by public organizations. Their multifaceted advantages encompass various aspects of responsible AI implementation:
- Data Protection and Privacy: SCCs provide a structured framework for safeguarding sensitive data and ensuring privacy compliance. By clearly defining data handling practices and responsibilities, they mitigate the risks of data breaches and misuse, aligning with the principles of data protection laws like the GDPR.
- Enhanced Transparency and Explainability: SCCs facilitate transparency in AI systems by requiring vendors to disclose their algorithms and data sources. This enhances the explainability of AI decision-making, enabling stakeholders to understand how and why AI systems reach particular outcomes.
- Liability and Accountability: One of the essential advantages of SCCs is their role in allocating liability and ensuring accountability. These clauses delineate responsibility in case of AI-related incidents, providing legal recourse and protecting both public organizations and the public interest.
- Termination and Exit Strategies: SCCs offer a clear roadmap for terminating contracts and exiting relationships with AI vendors. This is particularly valuable in high-risk scenarios, allowing public organizations to transition smoothly when required.
- Streamlined Procurement Process: SCCs provide a predefined structure, expediting the procurement process. This reduces the time and resources spent on negotiations and ensures that crucial ethical and legal considerations are addressed efficiently.
- Proven Success in Other Domains: SCCs have demonstrated their effectiveness in other sectors, including cloud computing and international data transfers. Case studies in these domains show how these clauses have improved compliance, data protection, and transparency.
- Long-Term Benefits: Beyond immediate advantages, SCCs promote ethical AI practices, fostering trust in public services. They contribute to the establishment of a responsible AI culture and enhance the long-term reputation and credibility of public organizations. Moreover, they align with international AI standards, making public procurement practices consistent with global expectations.
In an era where AI’s influence is ever-expanding, SCCs are indispensable tools in ensuring that public organizations uphold their ethical and legal responsibilities, protect sensitive data, and instill public trust in the deployment of high-risk AI systems. Their comprehensive advantages serve as a safeguard, promoting responsible and accountable AI practices across the public sector.
Conclusion
In conclusion, SCCs are pivotal in the responsible procurement of high-risk AI by public organizations. These clauses encompass data protection, transparency, liability, termination strategies, and streamlined procurement. Their adoption is crucial to ensure responsible AI deployment, protecting both public organizations and the public interest. Case studies across sectors demonstrate the positive impact of SCCs, and their long-term benefits include ethical AI practices, trust-building, and streamlined processes, making them a valuable asset for the public sector and the broader community.
