AI Drone Adoption Statistics: What PwC's 2022 UK Survey and 2024 Report Really Show

Is the PwC/ASSOCHAM 2024 report an original AI-drone adoption survey?

No. Opportunities in the drone sector with the adoption of AI and blockchain was published by PwC India and ASSOCHAM in October 2024 as a cross-sector report discussing drones, AI, geospatial analytics and blockchain. It includes sector opportunity figures and survey results reproduced from earlier research. The publication date of that overview must not be used as the collection date of the data it quotes.

Our original-source review identifies two different evidence chains. Survey chain: UK 2019 commercial-drone industry research, a 2022 follow-up survey published in Building Trust in Commercial Drones, and later reproduction in the 2024 India report. Market-opportunity chain: PwC's 2016 Clarity from Above based on 2015 estimates, repeated in 2018 Flying High, and reproduced as percentages in 2024. The separate historical 2017–2023 CAGR also comes through the 2018 report, which attributes that forecast to a third-party market-research source.

These are not interchangeable denominators. A 2022 UK opinion survey cannot quantify Indian AI-drone shipments in 2026; a 2015 addressable value of activities potentially replaced by drone solutions cannot be equated to revenue booked by commercial UAV manufacturers.

Who was surveyed, where and when?

The original PwC UK survey, About this survey (report page 20) gives the sample that the 2024 report omits from its summary tables. Opinium conducted the survey in 2022 among 251 people working across UK industries where drone services might be provided or purchased. They included 51 public-sector respondents (healthcare, government and public services) and 200 private-sector respondents. PwC states that 63 of the 251 had worked with commercial drones or purchased drone services. That is a count of respondents, entirely different from the 63% of survey respondents answering the separate credible-provider question. The main survey was not a sample of 251 active AI-drone purchasers.

A separate, smaller questionnaire sought qualitative reactions from selected drone-industry participants in late 2022 and early 2023. It obtained 18 responses: 10 from users/groups or organisations with drone teams, and eight from drone service providers. Do not use n=18 for the 63%, 88% or 43% broad industry survey results; they came from the 2022 n=251 research. PwC's 2019 baseline involved 252 senior UK business decision makers and a separate survey of 1,520 UK adults, described by PwC alongside its original 2019 public research; the 2022 report uses a similar industrial respondent profile to compare changes.

What do the 88%, 87%, 86% and 84% figures actually measure?

Table 2 of the 2024 report (printed page 11, PDF page 10) reproduces the perceived benefits data from the original 2022 UK commercial-drone study (Table 1). The figures express respondents agreeing that drones could benefit their industry, not achieved savings, production gains, flight-safety records or AI-system benchmark results.

Perceived benefits are a demand signal, not measured outcomes. For example, 88% agreement about potential time savings does not mean a drone or AI model reduced real project time by 88%. Similarly, the report's 43% value is the share of 2022 UK survey respondents who felt drones had been embraced by their industry and used effectively; it is not the proportion of all UK businesses with drones, and not a count of commercial AI integrations. The original 2019 comparator for that perception question was 33%.

The earlier UK study also notes that positive overall sentiment about commercial drones rose from 56% in 2019 to 72% in 2022. That 16-percentage-point shift is about attitudes among sampled industry respondents, not 16 percentage points of national drone or AI adoption. All figures should retain their original respondent base and time window when reused.

Why 63% is a supplier-credibility signal—not AI-drone penetration

Table 3 of the 2024 PwC/ASSOCHAM report compresses four UK adoption-barrier findings. The original 2022 PwC UK survey (Table 2) gives the original statement wording and the overall agreement shares:

Importantly, 63% agreed that more drone services would be used if more providers had credible offerings. It does not mean 63% of businesses had purchased an AI drone, that 63% of drone software vendors were credible, or that 63% of commercial drones had onboard AI. The 59% value describes the perceived gap in compelling evidence of benefits in the respondent's industry; the 56% relates to industry-specific qualifications.

The 44% was originally agreement that less restrictive regulations would increase drone use. The 2024 report shortens that to “reviewing the regulations,” which changes the nuance. Our interpretation follows the original survey question rather than treating the 2024 abbreviated table label as its precise wording. These historical barriers are useful inputs to B2B evidence marketing—buyers may seek supplier qualification, case studies and industry-specific outcomes—but the figures alone do not quantify demand for AI autonomy licences.

Where did the 36% infrastructure market share come from?

Table 1 in PwC/ASSOCHAM's 2024 report (printed page 10, PDF page 9) labels its percentages as the share of the drone market by industry. Tracing the same eight values to PwC's November 2018 Flying High report and the original May 2016 Clarity from Above study changes the interpretation: they derive from US$127.3 billion of potentially replaceable business services and labour measured using 2015 values. PwC calls it the total addressable value of drone-powered solutions, not recorded drone purchases, booked UAV revenue or sales of AI technology.

The original eight values total US$127.3 billion. Recalculated sector percentages are approximately 35.5% infrastructure and 25.5% agriculture, which were shown as rounded 36% and 25% in the later 2024 table. The eight published rounded percentages sum to 99%; the 1-percentage-point gap follows ordinary rounding, not a missing AI-drone industry segment. Even the more precise ratios are proportions of a historical addressable business-activity estimate, not independently measured annual drone-market sales.

It would be a material classification error to treat “infrastructure 36%” or “agriculture 25%” as shares of 2024/2026 AI-drone software, current unit sales or the US$127.3 billion as commercial UAV manufacturers' revenue. The 2016 study did not provide a verified AI feature penetration rate for those sectors.

Does PwC forecast 16–17% AI-drone growth in 2026?

No. The 2024 report repeats a 16–17% CAGR attributed to its November 2018 PwC Flying High report. The original 2018 wording explicitly states a global commercial-drone forecast for 2017–2023, citing Infinium Global Research as the underlying research provider. That forecast's target period had already ended before the 2024 publication, and it is neither a measured 2026 growth rate nor a generative-AI, navigation-AI or military-autonomy CAGR.

Do not compound this percentage to estimate 2026 AI-drone revenues. Such a forecast would require a separate 2020–2026 market series, clear geography and currency, software versus hardware scope, verified AI adoption, customer procurement observations, vehicle shipments and sources for each assumption. A historical forecast can be documented as context; it must not be represented as a current market fact.

What can UXV Searcher responsibly infer about AI-drone market demand?

The validated conclusions are qualitative: in 2022, sampled UK industry respondents considered commercial drones potentially beneficial, while reporting supplier-credibility, evidence and qualifications barriers. Such findings can guide research questions but do not establish how many AI-enabled systems are currently sold, the proportion of fleets with autonomy models, or an AI-specific revenue pool in India or worldwide.

For an evidence-led 2026 AI-in-drones market model, UXV Searcher would maintain separate records for actual customer and country, drone type and supplier, AI function, autonomous feature or service, software/compute/sensor spend, physical units, disclosed contract values, reported deliveries, source dates, evidence class and uncertainty. The model should subtract embedded component/software value where already counted in the aircraft price. Survey attitudes, addressable potential, historical forecasts and observed transactions require separate labels.

The practical marketing opportunity is to help buyers evaluate real suppliers, validated applications and deployment evidence—not to claim “63% AI drone penetration” or allocate 36% of an AI-drone TAM to infrastructure without an independently supported current denominator. Where no transaction-specific evidence exists, the appropriate status is NOT ESTABLISHED.