iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Councils are moving faster on AI, but the available evidence does not show that residents are already being left behind. The Local Government Association (LGA) found that 95% of responding English councils were using or exploring AI in its latest survey, up 10 percentage points from 2024. But the survey covered responses from only a third of councils, and the LGA says it is a snapshot rather than a representative picture of every council. Its figures also combine early exploration with operational use.
Are councils accelerating and leaving citizens behind?
The evidence supports a qualified answer: AI activity is increasing among the English councils that responded to the LGA survey, while their levels of readiness vary. It does not establish that councils across England are all accelerating at the same rate, or that AI adoption has caused residents to lose access to services.
The LGA conducted its survey from December 2024 to February 2025 and published the 2025 update on 2 June 2025. Because the results reflect responses from a third of councils, the adoption figures below describe respondents, not a census of local government. “Using or exploring” also covers different stages, from initial investigation to some use; it is not a measure of widespread or sustained deployment.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWhat are councils using AI for?
Among councils using or exploring AI, generative AI was the most commonly reported type. Respondents described applications in both internal administration and public services, so the figures should not be read as evidence that every example is operational or directly affects residents.
| Survey measure | Share of responding councils |
|---|---|
| Using or exploring AI | 95% |
| Generative AI | 83% |
| Perceptive AI | 28% |
| Predictive AI | 20% |
These LGA figures are from the 2025 survey update. The AI-type figures refer to respondents using or exploring AI, rather than all councils.
Readiness is not the same as adoption
The LGA’s account shows a wide range of maturity among respondents: half were at the beginning of their AI journey, 22% were developing capacity and capabilities, 14% were making some use, and 7% were described as innovative leaders. The stages are best read as evidence of uneven progress, not as proof that most councils have integrated AI into core services.
Uses range from back-office tasks to care
Common corporate uses reported by the LGA included HR, meeting administration, procurement, finance and cyber security. Respondents also reported AI use or exploration in adult social care (44%) and children’s social care (31%). These figures do not say that each application makes decisions about an individual, or that it is routinely used in a resident-facing service.
Rank #2
Just over two-thirds of respondents using or exploring AI were paying external suppliers or procuring AI tools or technologies. That describes councils’ institutional purchasing activity; it does not mean residents need to purchase a product to access council services.
What benefits and risks do councils report?
Respondents reported benefits including greater staff productivity, service efficiencies and cost savings. These are reported outcomes, not independently measured estimates of causal impact or net savings.
| Reported benefit | Share of LGA survey respondents |
|---|---|
| Staff productivity | 36% |
| Service efficiencies | 33% |
| Cost savings | 21% |
The figures are from the LGA’s 2025 survey update. The LGA also reported that many respondents saw opportunities in corporate functions, adult social care, and advice and benefits.
Rank #3
Funding, skills and capacity are common obstacles
In the same survey, the leading reported barriers were lack of funding (62%), staff capabilities (56%) and staff capacity (52%). Respondents using or procuring external tools also commonly described difficulties with scoping projects, understanding how AI is embedded in a product, and evaluating solutions. Together, these obstacles help explain why high interest should not be mistaken for consistent delivery.
Risk controls are not a guarantee
Respondents most often identified cyber security (83%), deepfake disinformation (69%) and organisational reputation and resident trust (68%) as significant risks. On managing AI risk, 49% said they used existing policies, 41% had a specific AI policy, and 38% used a Senior Responsible Owner. Those responses describe reported arrangements; by themselves, they do not demonstrate that controls are adequate or effective.
Why can responsible adoption be difficult?
In a qualitative account published in July 2026, the Ministry of Housing, Communities and Local Government (MHCLG) Digital team described April engagement sessions involving 60 councils. Participants reported varied stages of adoption, no clear shared minimum expectations for ethics, guidance that could be fragmented and difficult to apply, and gaps in practical skills and training for procuring, using and governing AI. The department said councils wanted practical guidance, usable tools and opportunities to learn from peers. These sessions offer insight into challenges participants raised; they do not estimate how prevalent each issue is across all councils.
MHCLG Digital’s AI Ethics Lead, Nimshi Mahamalimage, put one concern plainly: “When it comes to ethics, there are no clear minimum expectations that all councils should follow.” The LGA’s guidance describes responsible adoption as requiring strong data foundations, privacy protection, attention to bias, support for digitally excluded residents and staff, transparency, governance and oversight.
Those principles can be turned into practical questions for any proposed use:
Recommended Free Tools
- Purpose and impact: What service problem is the system meant to solve, and who is expected to benefit?
- Data and privacy: What information will be used, how will it be protected, and are the underlying data fit for the task?
- Bias and errors: How will the council look for unequal outcomes or mistakes, and how can a person challenge or correct an error?
- Human responsibility: Who oversees the system and remains accountable for decisions and service outcomes?
- Access and support: Can residents use the service without relying on AI or having strong digital skills, and is human help available?
- Evaluation: How will the council measure whether the service works for different groups, rather than simply counting time saved or tools purchased?
Are residents being left behind?
That is a legitimate concern to test, but the sources available do not establish it as a consequence of council AI adoption. National figures show that digital access and confidence vary, while neither the national statistics nor the LGA council survey tracks whether particular residents can successfully use AI-enabled council services.
The Department for Science, Innovation and Technology’s (DSIT) 2025/2026 public engagement survey found that 95% of UK adults had used the internet in the preceding three months. The figure was 78% for people aged 75–84 and 46% for those aged 85 and over. While 76% of adults said they could complete most digital tasks without difficulty, that fell to 13% among people aged 85 or over. These are population-wide indicators, not measures of access to a specific council service.
The same DSIT survey found that 59% of UK adults had used generative AI in the preceding three months, compared with 18% of people aged 75–84 and 8% of those aged 85 and over. Among adults who had not used generative AI, reported reasons included preferring to do things without AI (44%), privacy or security concerns (33%), not knowing enough to use it (32%), and concern about accuracy (31%). DSIT asked these reason questions online, a qualification that matters when interpreting responses about non-use.
The LGA’s Digital Inclusion Hub also highlights that digital exclusion can involve people described as “Not for Me” or “Reliant on Others,” not just people without an internet connection. It points to person-centred approaches and support from proxy users. In a separate Digital Communities Survey conducted in September and October 2025, 83 local authorities responded, a 26% response rate. This material broadens the ways to think about inclusion, but it does not measure the impact of AI-enabled council services on those groups.
What evidence would show whether residents are being left behind?
To answer that question, councils would need service-level evidence connecting adoption to residents’ actual experiences and outcomes. Useful measures would include:
- Which routes people can use to access the service, including non-digital options and access to human assistance.
- Whether interfaces and processes are accessible, and who can complete them without help.
- Error, correction and appeal rates, including whether residents can reach a human decision-maker.
- Service outcomes and waiting times compared with the service before AI was introduced.
- Differences in access, errors and outcomes by age, disability, language, income, connectivity and reliance on assistance.
- How privacy, bias, transparency and human oversight are handled in practice.
The LGA’s adoption survey, MHCLG Digital’s council engagement and DSIT’s public statistics illuminate separate parts of the issue. They do not provide this joined-up, service-level picture. Without it, claims that residents are already being left behind go beyond what these sources demonstrate; the same evidence also cannot rule out exclusion in particular services.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
