STOCKHOLM, Nov. 12, 2026 — AI errors are increasing business costs and delaying projects, according to new research from Paligo, a provider of governed context infrastructure for enterprise AI. The study found that 14% of respondents experienced direct financial losses or revenue impacts from incorrect or unreliable AI outputs, while 22% reported internal process or project delays and 18% incurred additional technology or computing costs.
The findings cover respondents’ experiences over the previous 12 months. At least 78 of the 411 respondents (19%) reported consequences involving safety, compliance or reputation. Across the full sample, 49% reported at least one of these impacts, including safety incidents or increased safety risks (17%), compliance, legal or regulatory issues (17%), and reputational impacts (15%).
The international study surveyed 411 business leaders across the UK, US, Germany and Benelux who were involved in enterprise AI decisions or the information used by AI systems. The findings are published in Paligo’s new report, What Does It Cost an Organization When AI Gets the Answer Wrong?.
AI Errors Create Measurable Business Costs
Rahul Yadav, CEO of Paligo, said: “Most AI errors aren’t model problems. They’re content problems. When an AI pulls an outdated manual or the wrong product version, it gives a wrong answer in a confident voice.
“Every wrong answer has a bill attached, and most organisations aren’t reading it. Half of those who hit errors spend five or more employee hours putting each one right, yet only around a quarter measure the time spent checking answers. That’s a potential context tax, paid quietly every day.”
Seven in ten respondents (70%) reported that an enterprise AI application in their organisation produced a materially incorrect answer or action at least monthly. However, only 15.6% said their organisation measures the cost of obtaining a reliable or verified answer. One in five (20.4%) measures cost per AI request, while 16.1% tracks token consumption.
Some AI services charge based on input and output tokens, which measure the amount of information a model reads and generates. Additional requests or unnecessary information can increase these costs. Nearly three-quarters of respondents (74.2%) said their systems frequently or occasionally retrieve the same or similar information. Processing repeated information can increase token usage, although repeat checks may sometimes be necessary to verify an answer.
Information Quality and Source Accuracy Remain Key Challenges
Problems with the information used by AI systems can make reliable outputs more difficult to achieve. Around 27% of respondents said investigations had identified outdated retrieved information, while 25% identified cases where relevant information from multiple sources had been incorrectly combined.
Among applications in production or near production, only 22% of respondents said their organisation could always identify which version of a source had been used.
Rahul Yadav continued: “Finding relevant information is only the beginning. A document can be accurate for one product and wrong for another. Before leaders trust an AI answer, they need to know which version it came from, whether it was approved and why it applied. If you can’t answer that, you can’t defend the answer.”
Verification and Correction Consume Employee Time
The cost of AI errors also extends to verification and correction activities. Among the 334 respondents who encountered AI errors, 51% reported that a typical issue required at least five employee hours in total to investigate, verify and correct.
Although 83% of all respondents reported frequently or occasionally checking AI outputs against original sources, only 27% said their organisation measures the time spent on verification.
Content-related problems are also affecting AI deployment timelines. More than three in five respondents (61%) reported frequent or occasional delays in deploying or expanding AI because supporting information was not sufficiently current, approved, structured or governed.
“The fastest way to more reliable AI is to fix the content at the source,” Rahul Yadav added.
“Make it current, approved and traceable, and have the AI flag a question or decline when no approved answer exists. Start with one product or one high-risk answer, and measure what changes.”
Paligo recommends tracking token usage and charges across every request required to obtain a usable answer, alongside employee time spent verifying and correcting outputs. Where missing context or conflicting source versions lead to repeated requests, organisations can assess whether improvements in source content or information retrieval reduce costs while maintaining answer reliability.
Read the full report: What Does It Cost an Organization When AI Gets the Answer Wrong?
Notes to Editors
For more information about Paligo, contact Stephen Sanders, Kiran Patel or Adam Gooseman at PR Agency One.
Telephone: +44 161 871 9140
Email: [email protected]
Methodology
The research was commissioned by Paligo and conducted in October 2026 among 411 individuals across the UK, US, Germany and Benelux who were involved in decisions about enterprise AI or the information it uses. Respondents represented organisations at experimentation, limited production and production stages across multiple functions.
The 70% figure represents respondents who reported materially incorrect AI outputs at least monthly during the previous 12 months. Correction-time findings cover 334 respondents who encountered errors, including those experiencing them less than monthly. Employee hours represent the combined time spent by staff investigating, verifying and correcting an issue.
Respondents could select multiple causes and consequences. The minimum of 78 people reporting safety, compliance or reputational impacts counts the largest of the three groups in each market: 26 in the UK, 21 in the US, 20 in Germany and 11 in Benelux. Token economics figures are presented to one decimal place, while other survey percentages are rounded to the nearest whole number.
The report describes a potential “context tax” associated with additional verification and information retrieval. The survey did not quantify token waste, potential savings, overall financial costs or AI’s net impact on productivity.
About Paligo
Paligo provides governed context infrastructure for enterprise AI. Organisations in manufacturing, medtech and other regulated industries use Paligo to keep documentation and operational knowledge structured, current and approved, enabling employees and AI systems to work with reliable information.
Founded in 2015, Paligo serves enterprises where inaccurate information can affect safety, compliance or production.
For more information, visit paligo.net.
