82% of Firms Report Positive AI Impacts Amid Data Risks
A Gallagher survey of over 1,200 global businesses found that 82% of respondents report positive impacts from AI adoption. However, data protection and the potential for errors remain the top challenges. The results indicate that while companies are increasingly embracing AI, they are also grappling with the associated risks.
- The Gallagher survey highlights that the primary barriers to AI adoption are a shortage of skilled talent (30%), ethical and data privacy concerns (30%), and complex compliance and regulatory issues (27%). A significant 85% of businesses surveyed have implemented job protection strategies to retain employees, with a key motivation being the preservation of creativity and the "human touch" in customer service. - For platform engineering teams, AI is being productized to create intelligent self-service platforms. These platforms use natural language processing, allowing developers to describe infrastructure needs conversationally, which AI then translates into provisioned resources. This is part of a larger trend where AI is not just assisting with coding but orchestrating entire developer workflows. - In the shipping and logistics sector, AI is optimizing operations by analyzing vast datasets, including weather patterns, ocean currents, and port congestion, to determine the most efficient routes. This leads to reduced fuel consumption and transit times. On the ground, AI-powered robots are integrated with warehouse management systems to automate picking, packing, and sorting, which reduces human error and scales operations during demand surges. - From a technical leadership perspective, integrating AI requires a robust tech stack that includes LLMs, RAG pipelines for grounding responses in company data, AI orchestration tools, and vector databases. Engineering leaders are finding it essential to establish comprehensive AI governance frameworks and focus on scalable architecture to handle increasing data volumes and user loads without performance degradation. - The rise of "Shadow AI"—the unsanctioned use of AI tools by employees—presents a significant data security risk. Inputting sensitive client or business data into third-party generative AI tools can breach confidentiality agreements and data protection laws like GDPR. Generative AI can also unintentionally reproduce fragments of confidential data from its training inputs, creating disclosure risks. - For developers, AI tools are shifting from assistive (autocomplete) to agentic, where AI proactively automates and orchestrates workflows across the entire software development lifecycle. AI-powered tools like GitHub Copilot and Amazon Q Developer can generate Terraform configurations, Kubernetes manifests, and deployment scripts, significantly accelerating infrastructure as code (IaC) development. - From a management perspective, engineering leaders are focusing on "curious scaling," which involves creating a culture where experimentation with AI is encouraged within defined guardrails. A key challenge is ensuring that AI augments focus rather than creating more noise, which requires providing the right support structures alongside new tools. Research shows that teams using AI-enhanced analytics have identified 42% more potential delivery risks than those using standard reporting. - The surge in AI is creating an API boom, with AI-related API calls increasing by 73%. This is driving the need for automated API management solutions that can handle documentation, intelligent traffic scaling, and security. AI is being used to analyze API traffic patterns to detect anomalies and predict usage spikes, shifting security from a reactive to a proactive stance.
