Accenture is under scrutiny following a report by GPTZero, which identified 15 false citations and fabricated case studies in the company's research paper titled Leapfrogging Development. The report, produced in collaboration with IE University in Madrid, explored the impact of artificial intelligence on agricultural productivity in Africa.
An Accenture spokesperson disputed the findings, attributing the errors to human oversight rather than AI hallucinations. The company is currently revising the document and plans to republish it. This incident follows previous warnings from Accenture CEO Julie Sweet, who stated that consultants failing to adopt AI could face termination as part of the firm's digital transformation.
The firm recently implemented new restrictions on AI tool usage after some employees generated monthly bills exceeding $1,000 using services like Claude. These internal controls reflect broader challenges in the consulting sector, where firms like Deloitte, EY, KPMG, and PwC have also faced allegations of AI-related inaccuracies in their reports.
Legal firms are also grappling with similar issues. Pinsent Masons in the UK admitted to including AI-generated hallucinations in documents submitted to the High Court. Additionally, Sullivan and Cromwell, which charges up to $3,000 per hour for legal services, acknowledged that AI errors appeared in filings submitted to a New York court earlier this year.
