The AI Tendering Paradox: Could Cheaper Bid Writing Make Public Procurement More Expensive?

A local authority publishes a major adult social care tender. Within weeks, its procurement team receives dozens of professionally presented submissions, each describing robust safeguarding, effective governance, personalised support and ambitious workforce development. The responses appear convincing, but how much of the promised delivery is supported by genuine operational evidence? And what happens if artificial intelligence makes producing such submissions substantially easier for every potential bidder?

This is the emerging AI tendering paradox. Technology that reduces the cost of preparing bids could, under certain conditions, increase the work required to evaluate them. The question reaches beyond tender-writing productivity into adult social care procurement strategy, tender quality and commissioning practice. It also challenges assumptions about how contracting authorities distinguish persuasive submissions from credible service-delivery commitments.

The development has implications for public procurement processes and law and the growing use of artificial intelligence and automation in care organisations. Neither widespread AI adoption nor increasing submission volumes should be assumed without evidence. Nevertheless, the underlying economic incentives are sufficiently significant to warrant serious examination.

For commissioners, the issue is how to maintain proportionate, fair and defensible evaluation. For providers, it is how to demonstrate genuine operational capability when fluent tender writing becomes increasingly accessible. For people receiving support, the consequences ultimately concern whether procurement decisions produce safe, sustainable and person-centred services.

Why AI Is Changing the Economics of Tender Participation

Historically, preparing a substantial adult social care tender could require significant investment in specialist writing, operational consultation, evidence gathering and internal review. Smaller providers sometimes struggled to participate because the cost of preparing a competitive submission was substantial relative to the probability of securing work.

Generative AI changes part of this equation. It can assist with interpreting questions, organising information, drafting narrative, checking consistency and adapting existing material. Activities that previously required considerable manual writing time can sometimes be completed more quickly, particularly where the underlying information is already available and reliable.

This creates genuine opportunities. Smaller organisations may be better able to participate in procurement exercises. Experienced providers with limited administrative capacity can devote more time to developing their service proposals. Procurement teams may receive clearer submissions from organisations that previously struggled to explain their strengths.

However, reduced production costs can also alter bidder behaviour. An organisation that previously selected two carefully matched opportunities might decide to pursue five or six because preparing the initial narrative appears less expensive. If many organisations make similar decisions, the aggregate number of submissions could rise even where the underlying number of viable providers remains unchanged.

The distinction between the cost of producing a submission and the cost of delivering a contract is fundamental. AI can reduce writing effort, but it does not automatically create additional Registered Manager capacity, specialist support workers, suitable accommodation, clinical partnerships or financial resilience.

Strong tender strategy and opportunity selection therefore remain essential. The ability to submit more bids is not necessarily evidence of an organisation's ability to deliver more services.

Could Lower Bid Costs Increase Commissioner Workloads?

The central economic concern is that savings achieved by bidders may generate additional evaluation work elsewhere in the procurement system. A provider can use AI to prepare a response more efficiently, but a contracting authority remains responsible for applying the published evaluation approach fairly and consistently.

Consider a procurement that previously attracted 12 submissions and subsequently attracts 35. If each submission contains several substantial quality responses, the additional evaluation workload could be significant. Evaluators may need to assess service models, scrutinise staffing proposals, compare mobilisation commitments and document their reasoning for each score.

This does not mean every additional bid creates an identical cost. Procurement systems, evaluation design, automated administrative processes and the quality of submissions all influence workload. Nor is there currently sufficient evidence to conclude that AI has produced a uniform increase in adult social care tender volumes across England.

The concern is instead a plausible structural risk: as the marginal cost of producing a bid falls, the cost of assessing the resulting competition may not fall at the same rate.

Evaluation also involves more than reading responses. Contracting authorities need appropriately skilled evaluators, moderation arrangements, records of decisions, conflict-of-interest controls and processes for addressing inconsistencies. Specialist services may require input from operational commissioners, safeguarding professionals, finance colleagues, clinicians or people with lived experience.

These demands become particularly important where a tender involves complex supported living, specialist mental health provision, acquired brain injury services or high-value extra care contracts. The evaluation task is not simply to identify the most polished response. It is to establish which tender best satisfies the published requirements and award criteria.

As digital procurement and contract management develop, contracting authorities may increasingly examine how administrative technology can support evaluation. Any such use must preserve transparency, fairness, information security and accountable human decision-making.

The Difference Between Persuasive Writing and Deliverable Care

One of the most important risks is the growing separation between the quality of written submissions and the maturity of the organisations submitting them.

Generative AI can produce convincing descriptions of safeguarding cultures, reflective supervision, positive behaviour support, multidisciplinary collaboration and continuous improvement. These descriptions may resemble the language used by experienced operational leaders even where the organisation has limited evidence of implementing the arrangements.

This is not a problem unique to AI. Tender submissions have always been capable of containing exaggerated promises, generic assurances and commitments that are difficult to verify. AI may, however, make producing such material faster and more accessible.

The operational distinction becomes apparent when examining the evidence behind a commitment. A statement promising monthly safeguarding audits has limited value unless the provider can explain who undertakes them, what information is reviewed, how concerns are escalated, how action completion is verified and what happens when recurring weaknesses are identified.

Similarly, a promise of person-centred support should be reflected in assessment methods, communication arrangements, staffing competence, decision-making processes and meaningful involvement of people receiving services.

Providers can use the Commissioner Evidence Builder to organise the relationship between tender commitments, supporting evidence, operational responsibilities and future contract assurance. The value lies in testing whether a proposed approach can be substantiated, not simply improving how it is described.

This creates an important challenge for bid-writing principles and evidence-led submissions. Strong writing remains valuable, but its credibility increasingly depends on the underlying service architecture and the accuracy of the commitments being made.

Scenario: A Domiciliary Care Framework Attracts More Submissions

Consider an illustrative English local authority renewing a domiciliary care framework. Its previous procurement attracted 18 bids, but the new exercise receives 47. Several smaller providers explain that AI-assisted writing has enabled them to participate without commissioning extensive external support.

The increase is not inherently negative. Greater competition may widen provider choice, improve market resilience and create opportunities for organisations with strong local knowledge. The authority must nevertheless evaluate the submissions according to its published requirements, including any applicable participation conditions and award criteria.

Many responses describe similar approaches to electronic care planning, safeguarding, staff supervision and continuity of care. Evaluators find that the language is often more sophisticated than the evidence supporting it. Some providers promise extensive weekend capacity without demonstrating realistic recruitment or rota assumptions.

The commissioning team cannot simply reject submissions because they appear AI-generated. Instead, it examines the published criteria, checks the evidence legitimately available within the procurement process and records how each response satisfies the requirements. Where clarification is permissible, it is managed consistently and without allowing an unfair improvement to the original tender.

The authority also reviews whether future procurement exercises could use clearer capacity requirements, more proportionate participation conditions and better-defined evidence expectations. Such changes must remain lawful, accessible and appropriate to the contract.

The lesson is not that increased competition is undesirable. It is that procurement design must distinguish genuine provider capability from the ability to produce a convincing written response. For people awaiting homecare, the ultimate test is whether commissioned capacity translates into reliable visits, continuity, safe practice and responsive support.

Procurement Law, Fair Competition and the Limits of Verification

In England, the Procurement Act 2023 provides the principal framework for many new public procurements commenced under the regime from 24 February 2025. Transitional arrangements mean that procurements commenced under earlier legislation can remain subject to the previous rules. Certain healthcare services may instead fall within the Provider Selection Regime, depending on the services and contracting arrangements involved.

This distinction matters because commissioners cannot redesign evaluation methods informally once submissions arrive. Award criteria, assessment methodology, participation conditions and relevant procedural requirements must be considered within the applicable legal framework.

The Procurement Act 2023 includes objectives relating to value for money, maximising public benefit, transparency and acting with integrity. Contracting authorities must also have regard to the importance of treating suppliers the same unless a difference justifies different treatment.

These principles become especially important when evaluators suspect that a submission contains generic AI-generated material. The lawful response is not to penalise a bidder merely for using technology unless the procurement documents and applicable rules provide a legitimate basis for doing so.

Instead, evaluation should focus on whether the tender meets the published requirements, how the evidence supports the commitments and whether any permitted clarification or verification is appropriate. Contracting authorities should obtain procurement or legal advice where the circumstances raise complex questions about misleading information, exclusion, clarification or procedural fairness.

Providers also need to understand that AI assistance does not transfer responsibility for their submissions. Directors and authorised signatories remain accountable for the accuracy of representations made on behalf of the organisation.

Further analysis of the Procurement Act 2023 and its governance implications is particularly relevant where organisations are reconsidering tender design, evaluation procedures or supplier assurance.

Designing Evaluation Around Evidence Rather Than Fluency

If AI-assisted drafting becomes routine, procurement evaluation may need to place greater emphasis on the relationship between a provider's proposed arrangements and its demonstrated capability. This does not require commissioners to abandon written quality responses. It requires questions that reveal how services will operate, how decisions will be made and how delivery risks will be controlled.

A broad question asking bidders to describe their commitment to safeguarding may generate relatively similar narratives. A more discriminating question could ask how safeguarding intelligence from multiple services is analysed, how recurring themes reach senior leaders and how changes to frontline practice are verified. The latter requires the bidder to connect policy, management systems and actual operational decisions.

Questions should remain proportionate to the service, the contract and the published evaluation framework. Smaller providers should not be disadvantaged simply because they lack elaborate corporate documentation. A well-managed local provider may demonstrate credible assurance through concise records, direct leadership oversight and practical examples rather than extensive committee structures.

Useful evaluation design may distinguish between several types of evidence:

• Capability evidence: relevant experience, resources, specialist competence and realistic delivery arrangements.
• Implementation evidence: how policies translate into supervision, staffing, assessment and operational controls.
• Outcome evidence: whether people experience meaningful improvements, continuity and appropriate support.
• Governance evidence: how risks, exceptions and recurring concerns are identified, escalated and resolved.
• Forward commitments: what the provider will establish during mobilisation and how progress will be verified.

The final distinction is particularly important for new providers. They may have credible plans without extensive historical evidence for a service they have not yet operated. Commissioners should distinguish legitimate future commitments from unsupported claims about arrangements supposedly already in place.

Effective tender review and evaluation therefore depend on the quality of the questions as much as the sophistication of the answers.

Scenario: An Extra Care Re-Procurement Where Existing Services Are at Stake

An illustrative housing-with-care provider is bidding to retain several extra care schemes. The contract represents a substantial proportion of its local operations and supports people who have developed long-standing relationships with familiar care workers.

The provider uses AI to organise the tender requirements and prepare initial drafts. However, its operational director recognises that the existing service presents complexities that generic narrative cannot adequately capture. Residents have different levels of need, staffing patterns vary between schemes and emergency response arrangements depend on the interaction between care staff, housing colleagues and external services.

Rather than accepting the generated narrative, the bid team works with Registered Managers, scheme leads, workforce colleagues and people receiving support. It tests proposed staffing commitments against actual rotas, reviews incident and safeguarding themes, and examines whether its mobilisation assumptions remain credible if the contract is awarded on revised terms.

The resulting submission is not necessarily the longest. It explains how existing continuity will be protected, how identified weaknesses will be addressed and how governance arrangements will support consistent practice across the schemes.

For the commissioner, the evaluation challenge is to recognise credible evidence without unfairly favouring the incumbent. Other bidders must have an equitable opportunity to demonstrate how they would deliver the service, including their arrangements for transition, recruitment, resident communication and risk management.

For residents, the procurement outcome is not an abstract commercial decision. It can influence who supports them, how changes are communicated and whether their independence, relationships and routines are protected. This is why the quality of procurement assurance matters beyond the written scores.

Why CQC Evidence and Tender Evidence Must Be Connected

In England, providers carrying on regulated activities are subject to the Health and Social Care Act 2008 regulatory framework, including the Health and Social Care Act 2008 (Regulated Activities) Regulations 2014. Relevant requirements include person-centred care, safe care and treatment, safeguarding, staffing and good governance.

These requirements are not tender-writing standards. Nevertheless, they provide an important foundation for assessing whether a provider's proposed arrangements are credible. A submission promising effective governance should be consistent with the systems through which the organisation actually monitors quality, manages risks and improves services.

CQC's assessment approach examines evidence relevant to quality statements, including governance, management and sustainability, safe systems and pathways, learning culture and involvement of people. The relevance of individual statements depends on the regulated activity and the circumstances being assessed.

A commissioner may legitimately examine similar operational themes, but local authority contract evaluation and CQC regulatory assessment remain distinct functions. A high-scoring tender does not establish regulatory compliance, and a favourable CQC rating does not automatically demonstrate that a provider can deliver a different or substantially expanded service.

Providers can use the CQC Evidence Gap Analyzer to examine whether their regulatory evidence is sufficiently developed and whether there are inconsistencies between documented arrangements and operational assurance. Such analysis can inform bid preparation without replacing regulatory judgement or contract-specific evidence.

The relationship between CQC evidence and provider assurance becomes especially significant where tender commitments involve new specialist services, rapid expansion or complex workforce arrangements.

A mature provider should be able to trace a central tender commitment through the relevant policy, accountable manager, workforce competence, operational record and governance reporting process. Where the commitment concerns a future service, the provider should explain how these elements will be established and tested during mobilisation.

Scenario: A Complex Supported Living Tender Tests Specialist Capability

A local authority commissions a new supported living service for adults with learning disabilities and complex support requirements. The specification emphasises communication, positive behaviour support, community participation, safeguarding and individualised staffing arrangements.

Several bidders submit well-written descriptions of person-centred care. One response promises highly individualised support but offers limited explanation of how specialist competencies will be maintained during staff absence, how distress will be understood or how restrictive interventions will be reviewed.

Another bidder provides a more developed operational model. It explains how individual communication profiles influence recruitment and induction, how behavioural assessments inform support planning and how specialist supervision is linked to direct observation of practice. Its mobilisation proposal distinguishes essential pre-opening competencies from development activities that can safely continue after commencement.

The commissioner evaluates both submissions against the published criteria, considering the relevance and credibility of the evidence rather than the apparent sophistication of the language.

For the second provider, its proposed supported living governance and assurance arrangements also connect operational oversight with the rights and preferences of the people who will use the service.

The distinction matters because specialist support cannot be reduced to a set of attractive commitments. A person who communicates distress through changes in behaviour may depend on staff recognising subtle indicators, understanding environmental triggers and responding consistently. Weak implementation could undermine wellbeing, independence and trust even where the original tender response appeared convincing.

The strongest procurement evidence therefore shows how specialist knowledge will influence everyday decisions, how practice will be supervised and how the provider will identify whether its approach is improving people's lives.

Governance Accountability When AI Contributes to a Bid

For providers, responsible AI-assisted tendering begins with clear accountability. AI may help draft, analyse or organise information, but the organisation remains responsible for the accuracy and deliverability of its submission.

A mature approach identifies who approves the use of AI, what information can be entered into external systems, who verifies generated statements and how final commitments are authorised. The level of control should reflect the sensitivity of the material and the potential consequences of error.

Information governance is especially important in adult social care. Tender preparation can involve commercially sensitive pricing, employee information, service-user examples, safeguarding themes and confidential commissioner documents. Uploading such material into an unsuitable AI service may create privacy, contractual and security risks.

UK data protection requirements, including the UK GDPR and Data Protection Act 2018, remain relevant where personal data is processed. Organisations need an appropriate lawful basis, suitable security measures and clarity about how information is handled by technology suppliers. Anonymisation should be genuine rather than simply replacing a person's name while leaving them identifiable through other details.

The Digital Transformation Readiness Assessment offers a practical way to examine digital governance, workforce capability, supplier assurance and cyber resilience before expanding AI-supported workflows.

Within governance arrangements for tender submissions, accountability should extend beyond the bid writer. Operational leaders need to validate service models, finance colleagues need to confirm commercial assumptions and appropriate senior managers should authorise material commitments.

Boards and directors should receive assurance about material risks rather than routine drafting activity. For a major contract, this may include the realism of mobilisation, workforce availability, financial exposure, dependencies on partners and the consequences of unsuccessful delivery.

Scenario: An AI-Assisted Mobilisation Plan Conceals a Workforce Gap

A provider preparing a specialist community care tender uses generative AI to develop a detailed mobilisation programme. The resulting narrative promises recruitment, induction, specialist training, competency assessment and operational readiness within eight weeks.

During internal review, the workforce lead identifies that the proposed timetable assumes applicants can be recruited, appropriately checked and trained more quickly than recent experience supports. The service also requires staff competent in delegated healthcare activities, with oversight from relevant healthcare professionals.

The bid team revises the proposal using actual recruitment data, training capacity, supervision arrangements and contingency assumptions. It distinguishes activities that can run concurrently from those dependent on earlier recruitment or assessment stages.

The revised plan includes clear decision points at which senior leaders will review whether mobilisation remains achievable. If critical recruitment milestones are missed, the provider will escalate the issue, review safe alternatives and communicate with the commissioner rather than relying on an unrealistic original timetable.

This demonstrates why AI-generated plans require operational testing. A plausible sequence of activities is not necessarily a deliverable programme. The quality of the submission improves when assumptions are challenged before they become contractual commitments.

For people awaiting support, this scrutiny matters because an unrealistic mobilisation plan can lead to delayed services, inconsistent staffing or pressure to commence delivery before the workforce is adequately prepared.

Commissioning Costs Extend Beyond the Evaluation Stage

The economic consequences of procurement decisions are not limited to the time spent scoring submissions. An inadequately evidenced award can create additional costs during mobilisation, contract monitoring, remedial action or service transition.

For example, a provider may offer an ambitious staffing model that proves unaffordable once actual recruitment and travel costs become clear. A proposed digital reporting system may not integrate with existing processes. A specialist service may lack the supervision capacity required to maintain competence across its workforce.

These risks existed before generative AI. The concern is that increasingly polished submissions could make weaknesses harder to recognise unless evaluation is sufficiently focused on operational substance.

Commissioners therefore benefit from connecting procurement evaluation with subsequent contract assurance. Commitments made during tendering should be translated into appropriate mobilisation milestones, reporting arrangements, quality measures and escalation processes.

The Quality Dashboard Builder can help providers structure the operational information needed to monitor service performance, identify variation and report meaningful trends. Its usefulness depends on the reliability of the underlying data and the quality of leadership interpretation.

Effective quality assurance and auditing should test whether commitments are being implemented rather than simply confirming that documents exist.

The broader public value calculation therefore includes evaluation expenditure, mobilisation reliability, contract management capacity and the consequences of service disruption. A cheaper tender-writing process does not automatically produce a cheaper procurement system.

Could Procurement Processes Evolve to Manage AI-Generated Submissions?

The longer-term question is whether public procurement methods will adapt as AI-assisted tender production becomes more widespread. Several developments are plausible, although their adoption will depend on procurement law, contracting authority capacity, market conditions and the characteristics of individual services.

One possibility is greater emphasis on structured evidence rather than extended descriptive responses. Commissioners could ask more focused questions about operational responsibilities, implementation dependencies, workforce assumptions and measurable outcomes. This may improve evaluation efficiency while making it harder for unsupported narrative to substitute for credible delivery arrangements.

Another possibility is more sophisticated use of staged procurement processes where the applicable regime permits them. Proportionate participation conditions, preliminary market engagement and clearly defined evaluation requirements can help authorities identify relevant capability without creating unnecessary barriers for smaller providers.

Interviews, presentations and practical demonstrations may also become more important where they are appropriate to the procurement and included within the published evaluation approach. Such activities can allow evaluators to explore how operational leaders understand their proposals, although they are not immune to rehearsed answers or superficial presentation.

Technology could support commissioners through administrative automation, document organisation and analysis of evaluation information. However, AI-assisted evaluation introduces its own risks, including bias, inconsistent interpretation, confidentiality concerns and difficulty explaining automated outputs.

Government procurement guidance already recognises the potential benefits and risks of suppliers using AI in bids. Transparency about AI use, appropriate information controls and proportionate assurance are therefore relevant considerations. The existence of guidance should not be confused with a universal prohibition on AI-assisted tender writing or a general requirement for every contracting authority to adopt automated evaluation.

Future technology and digital innovation in tenders will be most valuable where it improves the quality of procurement decisions rather than simply accelerating document production.

Scenario: A Commissioner Redesigns a High-Value Care Procurement

An English local authority reviews its experience of procuring a substantial community-based care contract. The previous exercise generated extensive written submissions, with evaluators reporting difficulty distinguishing genuinely different operational approaches.

Before the next procurement, the authority undertakes preliminary market engagement and examines which aspects of the previous evaluation provided meaningful assurance. It identifies that questions about governance structures generated lengthy responses, while more focused questions about workforce deployment and service continuity revealed significant differences between bidders.

The authority develops a revised approach within the applicable procurement framework. Quality questions place greater emphasis on how providers would respond to realistic operational pressures. Evaluation guidance explains the evidence required, the relevant scoring considerations and the importance of credible implementation.

People with lived experience contribute to the development of relevant service outcomes and accessible engagement arrangements. Their involvement helps ensure that evaluation is not dominated by corporate language at the expense of continuity, dignity, choice and everyday independence.

The authority also reviews the resources required for evaluation and ensures that appropriate operational expertise is available. Rather than attempting to identify AI-generated text, it concentrates on whether submissions demonstrate the capabilities and commitments required by the contract.

Following award, relevant tender commitments are incorporated into mobilisation and contract monitoring arrangements. The authority can then examine whether the service being delivered reflects the promises on which the procurement decision was based.

This approach illustrates a potentially more sustainable response to AI-assisted bidding: improving the quality of procurement design and assurance rather than treating the technology itself as the principal problem.

The Future of Specialist Bid Writing and Operational Expertise

AI is likely to continue changing the economics of routine tender production. Activities involving formatting, summarising requirements, reorganising existing information and producing initial narrative drafts may become increasingly efficient.

This creates pressure on consultancy models that derive much of their value from the time required to produce written responses. Where clients believe that acceptable submissions can be generated at substantially lower cost, traditional pricing arrangements may become harder to sustain.

However, the value of specialist consultancy is not necessarily determined by writing speed. Complex adult social care procurements require judgement about operational feasibility, commissioner expectations, regulatory obligations, workforce design, financial sustainability and the consequences of service failure.

A consultant supporting a high-value contract re-procurement may contribute to service strategy, evidence validation, risk assessment, mobilisation planning and executive decision-making. These activities extend beyond drafting and cannot be reliably replaced by fluent text generation alone.

The distinction is especially important where providers are competing to retain substantial existing contracts or secure identifiable new service delivery. Losing such a procurement can affect organisational income, workforce stability and continuity of care. Winning a contract on the basis of unrealistic commitments can create equally serious problems.

Specialist expertise therefore becomes valuable when it helps an organisation make better decisions, identify weaknesses before submission and develop proposals that are both competitive and deliverable.

This does not mean human-written submissions are automatically superior to AI-assisted ones. A carefully governed AI-supported process may produce stronger work than an inadequately researched manual submission. The differentiating factor is the quality of evidence, operational understanding and professional challenge applied throughout development.

Organisations considering their future bid capability should examine the relationship between technology, internal expertise and external specialist support. The most effective model may combine AI-enabled administrative efficiency with experienced operational leadership, structured quality assurance and clear accountability.

Measuring Whether Procurement Is Becoming More Effective

As AI changes tender participation, commissioners and providers would benefit from better evidence about its actual effects. Claims that AI is increasing bid volumes, reducing quality or overwhelming evaluation teams should be tested rather than assumed.

Contracting authorities could examine changes in the number of submissions received, evaluator time, clarification activity, the proportion of compliant bids and the relationship between evaluation scores and subsequent contract performance. Such analysis would need to account for contract value, market conditions, procurement design and changes in the number of available suppliers.

Providers could examine their own bid selection decisions, preparation costs, success rates, evidence quality and the extent to which awarded contracts achieve the outcomes originally proposed.

Importantly, the strongest measures would extend beyond procurement administration. They would examine whether services commence safely, workforce commitments are achieved, people experience continuity and commissioners receive reliable evidence of outcomes.

These measures can also inform continuous improvement across provider organisations, helping leaders distinguish between efficiencies that genuinely strengthen delivery and those that simply reduce the cost of producing documentation.

The objective should not be to preserve traditional tender-writing processes for their own sake. It should be to ensure that new technology improves access, competition, decision-making and public value without weakening accountability.

Conclusion

The AI tendering paradox presents an important question for the future of adult social care procurement: could technology that makes bids cheaper to produce also increase the resources needed to evaluate them and verify their credibility?

The answer remains uncertain. AI may improve access to procurement opportunities, reduce unnecessary administration and help providers communicate their strengths more effectively. These are meaningful benefits, particularly for smaller organisations with limited bid-writing capacity.

At the same time, lower submission costs could encourage greater participation, increase evaluation workloads and make it more difficult to distinguish persuasive narrative from genuine operational capability. Where major care contracts are involved, the consequences extend beyond procurement efficiency into service continuity, workforce sustainability, safeguarding and the everyday experiences of people receiving support.

The strongest response is not to resist AI or assume that traditional tender writing is inherently more reliable. It is to strengthen the connection between procurement requirements, operational evidence, professional judgement and accountable service delivery.

Commissioners need evaluation approaches that are proportionate, legally defensible and capable of recognising credible implementation. Providers need governance systems that ensure tender commitments reflect what their organisations can realistically deliver. Specialist advisers need to demonstrate value through judgement, evidence and operational understanding rather than writing volume alone.

Ultimately, successful procurement should not be measured by how quickly a submission is produced or how convincingly it is written. It should be measured by whether the resulting decisions secure sustainable services, responsible use of public resources and better outcomes for the people whose lives depend on them.