Seventy-seven percent of global businesses are actively pursuing the use of artificial intelligence (AI). Yet, despite this widespread adoption, an astonishing 91% of organizations admit they are unprepared to manage the risks AI introduces.
Indusface, a leading application security SaaS company, has conducted research on UK businesses to identify the sectors most vulnerable to AI dependency. Venky Sundar, Founder and President of Indusface, highlights the potential dangers of over-reliance on AI in the business world.
A recent Forbes Advisor 2024 AI Report reveals which UK industries are at the forefront of AI adoption:
| Rank | Industry | Businesses Currently Using AI (%) | Businesses Planning to Use AI (%) |
|---|---|---|---|
| 1 | Legal | 30% | 31% |
| 2 | IT and Telecoms | 29.5% | 27% |
| 3 | Finance and Accounting | 26% | 20% |
| 4 | Media, Advertising, and Sales | 20% | 21% |
| 5 | Transport and Distribution | 17% | 16% |
| 6 | Manufacturing | 16.5% | 14.5% |
| 7 | Education | 15.5% | 15.5% |
| 8 | Construction | 12% | 15% |
| 9 | Real Estate | 12% | 14.5% |
| 10 | Hospitality and Leisure | 11% | 9% |
These industries are experiencing significant benefits from AI, but as reliance increases, so do the associated risks.
The legal sector stands out as one of the most prominent users of AI, transforming legal research, analysis, and due diligence. However, this rapid adoption prompts the question: is the sector moving too quickly? Nearly a third (30%) of UK law firms have integrated AI into their operations, significantly enhancing the quality of their services.
Following closely, the IT and telecoms sector sees 29.5% of businesses using AI, benefiting from applications such as network optimization, predictive maintenance, customer service automation, and security enhancements.
In the finance and accounting sector, more than a quarter (26%) of UK businesses are already leveraging AI for tasks like fraud detection, credit risk assessment, financial reporting, and customer service. But with these advancements come potential new vulnerabilities—could AI’s role in financial services expose businesses to unforeseen risks?
The Growing Concern of AI Dependency
While the advantages of AI in business are clear, there’s a growing concern about over-dependence on this technology. According to Forbes Advisor research, nearly six in ten (59%) Brits express concern about the increasing use of AI.
Venky Sundar shares his perspective on the risks of AI in business:
“Any sector where technology is a critical component, from farming to finance, is at risk of becoming overly reliant on AI. Sectors that are data-heavy and rely on analytics to remain competitive—like e-commerce, retail, credit, and risk rating—are likely to adopt AI more rapidly than others. However, it’s crucial to remember that proof of concept (POC) should remain just that—a concept. Rushing to market with a POC can have serious repercussions for application security and data privacy. AI-generated code often lacks secure coding practices, making it vulnerable to attacks.”
Sundar also warns of the risks associated with using large language models (LLMs) as input interfaces for products. “Traditional software inputs were predictable, controlled by form fields like text boxes or dropdown lists. But AI’s ability to interpret and respond to prompts removes this predictability, leading to potential security risks. Early signs of these ‘prompt injections’ are already being documented by the Open Worldwide Application Security Project (OWASP).”
Mitigating AI Risks in Business
To counter these risks, Sundar advises that AI should be seen as a “sidekick,” not a crutch—especially in software development. Before launching products, companies should adhere to best practices such as vulnerability scanning, penetration testing, patching, and implementing Web Application and API Protection (WAAP) measures to prevent attacks.
Given the threat of LLM injections, Sundar emphasizes the need for separate, continuously updated knowledge bases for productivity and defense use cases. He stresses the importance of human oversight and the necessity for diverse teams to maintain the datasets, ensuring they are robust and inclusive.
In summary, while AI offers transformative potential for businesses across various sectors, it’s imperative to approach its adoption with caution and a strong emphasis on security and ethical considerations.
4o