Artificial Intelligence is increasingly being used to automate fire engineering calculations and design. We assess the benefits against the risks of deskilling and error.. AI in Fire Engineering: A Tool for Enhanced Safety or a Major Risk? The burgeoning integration of Artificial Intelligence (AI) into fire engineering design and calculation processes is sparking both excitement and trepidation across the UK's built environment sector. As algorithms begin to automate complex tasks, from fire spread modelling to evacuation simulations, the promise of enhanced efficiency, accuracy, and ultimately, improved life safety, looms large. Yet, beneath this shiny veneer of technological advancement lies a crucial question: can we truly trust the algorithm with the lives of building occupants, or does this reliance introduce new, unforeseen risks of deskilling, algorithmic bias, and catastrophic error in a highly regulated and safety critical field? Background Fire engineering, by its very nature, is a discipline rooted in mitigating risk and safeguarding lives. Traditionally, it has relied heavily on the expertise, experience, and professional judgement of human engineers. This involves interpreting complex building regulations, such as those enshrined in Approved Document B (ADB) or the more prescriptive British Standards like BS 9991 and BS 9999, and applying fundamental fire science principles to design robust fire safety strategies. The Grenfell Tower tragedy, and the subsequent Hackitt Review, profoundly underscored the critical importance of competence and accountability in fire safety, leading to significant legislative reforms including the Building Safety Act 2022 (BSA 2022). This evolving regulatory landscape demands an even greater degree of rigour and demonstrable competence from fire safety professionals. The advent of AI presents a paradigm shift. Machine learning algorithms, deep neural networks, and generative AI are now capable of processing vast datasets, identifying patterns, and even generating design solutions at speeds and scales unimaginable to human engineers. Early applications in fire engineering have focused on automating repetitive calculations, optimising smoke control systems, predicting fire behaviour in complex geometries, and even assisting in the development of evacuation strategies. Proponents argue that AI can reduce human error, accelerate design cycles, and identify optimal solutions that might be overlooked by conventional methods. Key Developments Several key developments are driving AI adoption in fire engineering. Firstly, the increasing computational power available has made sophisticated simulations and data analysis more accessible. Secondly, the growing availability of fire related data – from sensor readings in smart buildings to historical fire incident data – provides the necessary fuel for AI algorithms to learn and improve. We are seeing AI powered tools emerge that can: Automate prescriptive compliance checks: AI can rapidly cross reference design proposals against the requirements of ADB or specific clauses within BS 9991/9999, flagging potential non compliances. Enhance fire modelling: Advanced computational fluid dynamics (CFD) models are being augmented with AI to predict smoke and heat spread with greater accuracy and speed, especially in complex, multi zone buildings. Optimise evacuation strategies: AI can analyse building layouts, occupant densities, and egress routes to simulate evacuation scenarios, identifying bottlenecks and suggesting improvements far more comprehensively than traditional manual methods. Support performance based design: For complex projects where prescriptive guidance falls short, AI can assist engineers in developing and validating performance based solutions, a critical aspect of modern fire engineering practice. Predictive maintenance: AI can analyse data from fire detection and alarm systems to predict potential failures, enabling proactive maintenance and reducing the risk of system downtime, a key factor in maintaining compliance with the Regulatory Reform (Fire Safety) Order 2005 (RRO 2005). However, these advancements are not without their caveats. The "black box" nature of some AI models, where the decision making process is opaque, poses a significant challenge. How can an engineer or a regulator scrutinise and validate a design recommendation if the underlying logic is inscrutable? This is particularly pertinent in the context of the BSA 2022, which places a strong emphasis on accountability and demonstrable competence throughout the building lifecycle, including the design phase. Regulatory Implications The UK's regulatory framework for fire safety is robust and continually evolving, driven by lessons learned and the imperative to protect life. The BSA 2022, with its focus on higher risk buildings (HRBs) and the establishment of the Building Safety Regulator (BSR), introduces stringent requirements for competence, golden thr