TL;DR

Source Global Research asked 3,887 companies what they make of consultants using AI, and 70% said they would not trust a report produced with it. Almost a third said learning a firm had used AI in a published report would shake their confidence in that firm. The uncomfortable detail: clients who have actually seen these tools working are more sceptical, not less.

That inversion is the finding worth sitting with. Chief executive Fiona Czerniawska noted the expectation had been the opposite — exposure ought to build confidence. Instead, 77% of those who have used consultants’ AI tools think the technology is a bubble about to burst, against 55% of those who have not.

The timing is poor for firms that have already committed. More than three-quarters of UK consultancies now use AI in delivery, according to the Management Consultancies Association. The big four have all cut graduate intake since ChatGPT launched while spending heavily on the technology, with KPMG and PwC striking deals with Microsoft and OpenAI, and PwC running its own ChatPwC assistant since 2023. The bet was that AI would substitute for junior labour. Clients are now saying they will not pay the same for the output.

The errors have not helped. GPTZero found several PwC Middle East reports containing invented footnotes and claims that could not be verified; PwC said it was correcting a limited number of citations and pointed to its quality control processes. Deloitte agreed to hand back part of its fee to the Australian government over a piece of work worth A$440,000, or roughly £230,000, whose mistakes were attributed to AI.

Two structural threats sit behind the reputational one. OpenAI and Anthropic have both launched consulting arms aimed at helping clients deploy AI, competing directly with the incumbents. And if AI genuinely delivers efficiency, clients will want the saving. KPMG demonstrated the logic on itself this year, pressing Grant Thornton UK, which audits it, for a lower fee on exactly that basis — the 2025 engagement came in at $357,000, about £265,000, down from $416,000 twelve months earlier.

For UK firms selling AI-augmented services of any kind, the read-across is direct. Disclosure is becoming a commercial decision rather than an ethical one, and the market is currently pricing AI involvement as a discount rather than a premium. Elixirr’s shares rose 9% yesterday on accelerating AI advisory work, which suggests advising on AI remains a better business than delivering with it.

Looking Forward

Watch whether any major firm starts publishing its AI usage policy to clients as a differentiator. The MCA’s line is that the value lies in combining the tools with human judgement — a position that only holds if clients can tell which is which.