“It’s Just a Wild, Wild West”: Harnessing Public Procurement as an AI Governance Mechanism
This is a strong CHI policy-and-practice paper: it reframes procurement as an AI governance lever and backs that claim with interviews from relevant stakeholders. The contribution is mainly a grounded synthesis of practices and mechanisms, not a technical system, so its impact depends on whether public buyers can operationalize the recommendations.
Axes Lens
Rare contribution shape, typical evidence profile. The point here is not a score. It is to show what kind of claim the paper makes, and whether the evidence pattern is unusual or baseline in this 268 -review set.
Contribution shape
- Knowledge form
- normative knowledge typical · 31/268
- Novelty type
- synthesis typical · 16/268
- Abstraction level
- practice typical · 85/268
- Generalization target
- organizational context typical · 20/268
- Validation mode
- qualitative study typical · 63/268
Evidence profile
- Evidence strength
- moderate typical · 105/268
- Claim alignment
- strong typical · 231/268
- Overclaim risk
- medium typical · 210/268
Review Summary
This paper is best read as a governance and practice contribution for public-sector AI, not as a technical AI paper. Its central insight is that procurement can be treated as a leverage point for shaping what AI gets adopted and under what conditions, which is a meaningful departure from the usual view of procurement as an administrative purchasing function. The abstract makes the contribution clear: the authors draw on semi-structured interviews with UK and EU buyers, providers, and procurement experts, and from that evidence derive six procurement practices plus concrete mechanisms to support uptake. That combination suggests a synthesis-style contribution with normative force: it does not merely describe a problem, but proposes actionable ways to align procurement with public interests. The paper also appears to add an important field diagnosis: AI-specific procurement approaches are still immature, and systems often enter through informal channels with less scrutiny. That is valuable because it explains why governance via procurement is hard in practice, not just why it is desirable. The evidence base is appropriate for the claim type: qualitative interviews can support a practice framework and reveal institutional dynamics, but they do not justify broad statistical generalization. The limitations language is therefore important and credible: the study is exploratory, based on a niche and difficult-to-reach expert population, and the findings are not statistically generalisable. I would rate the contribution as strong within its scope, with moderate evidence strength and medium overclaim risk only if readers treat the practices as universally applicable rather than context-sensitive guidance for UK/EU public procurement settings.
What Changed
Canon before
Public procurement is typically treated as an administrative purchasing process rather than as an explicit AI governance mechanism. The paper’s contribution is to reframe procurement as a governance lever and to show, from interviews with UK and EU procurement stakeholders, how that lever can be operationalised through concrete practices, mechanisms, and institutional supports.
Departure from common sense
The paper’s core move is to recast procurement as governance: “Recent work positions public procurement as a way to shape public sector AI in line with public interests, using the state’s purchasing power to influence which AI systems are procured and under what conditions.” That framing departs from the common-sense view of procurement as back-office buying.
Actual novelty
The paper contributes an empirically grounded set of six procurement practices, plus concrete mechanisms for uptake, and it also surfaces a field condition that AI-specific procurement remains immature and often bypassed through informal channels. The novelty is not a new algorithm but a practice-oriented governance synthesis for public-sector AI procurement.
Evidence
The abstract and main text show a qualitative interview study with UK and EU buyers, providers, and procurement experts. The paper’s contribution is a practice-oriented governance synthesis: six procurement practices, concrete mechanisms for implementation, and a diagnosis that AI procurement is still immature and often routed through informal channels. The evidence supports bounded, context-sensitive recommendations rather than broad statistical claims.
“ Our findings result in six promising procurement practices that enable the public sector to shape AI in line with public interests, alongside concrete mechanisms to support their uptake”
actual novelty · Abstract / contributions · confidence 0.70
“ It is envisaged that many such services can be mediated by AI and automated systems, and indeed, this is increasingly the case [ 155 ], raising challenges for their integration, often due to the complex stakeholder networks that characterise public sector services [ 83 , 99 , 100 , 139 ]. Increasingly, AI systems are introduced into high-risk”
departure from common sense · Background / public procurement as policy instrument · confidence 0.66
“ As a qualitative, exploratory study among a relatively niche and difficult-to-reach group of experts, the findings are not statistically generalisable [ 156 ] but are likely to offer useful insights from real-world practice”
limitation · Limitations and future work · confidence 0.92
“ This paper examines how this potential can be realised in practice by drawing on semi-structured interviews with UK and EU buyers, providers, and procurement experts”
validation scope · Abstract / methods scope · confidence 0.74
Limits
Method limits
As a qualitative, exploratory study among a relatively niche and difficult-to-reach group of experts, the findings are not statistically generalisable and should be read as practice-informed synthesis rather than population-level evidence.
Deployment limits
The recommendations are most directly applicable to UK and EU public-sector procurement settings, especially where buyers have formal contracting leverage and can influence tender criteria, framework use, and post-award monitoring.
Boundary conditions
The contribution is bounded by semi-structured interviews, by the institutional context of UK and EU procurement, and by the fact that the paper addresses procurement as one governance lever among others rather than a standalone solution.
Position in field
This sits at the intersection of HCI, public-sector AI governance, and procurement practice. Its value is in translating governance aspirations into procurement practices and mechanisms, rather than proposing a technical control layer.