RESEARCH
Grounded in doctoral research and real-world leadership experience, L-ARC™ explores how artificial intelligence, governance and institutional accountability can work together to support financially sustainable preventive action.
Research areas include:
About the Book
The Fire That Should Never Be asks a simple but important question:
Why do institutions so often recognise the value of prevention only after severe and avoidable damage has occurred?
Using wildfire and environmental risk as its central context, the book explores how artificial intelligence, governance and economic reasoning can support earlier and more responsible intervention.
Its central argument is that AI should do more than improve prediction. It should help decision-makers recognise emerging risk, understand its consequences and act before environmental threats become irreversible human, ecological or economic losses.
The book makes the case for a shift from reactive response to anticipatory prevention.
From Prediction to Prevention
Better prediction does not automatically produce better outcomes.
A warning has limited value if it is not understood, responsibility is unclear, decision-makers do not trust the evidence, or action cannot be mobilised in time.
The real challenge is therefore to strengthen the pathway:
Environmental Signal → Intelligence → Decision → Intervention → Resilience
The book examines how technology, human judgement, institutional responsibility and the economics of prevention can work together across this pathway.
What the Book Explores
Artificial Intelligence and Environmental Risk — how AI, environmental data and geospatial intelligence can support earlier detection and interpretation of risk.
Prevention Before Response — why institutions often invest heavily after disaster while underinvesting in preventive action.
Human Judgement and Governance — why consequential decisions still require accountability, context and human responsibility.
Economics of Prevention — how the cost of acting early compares with the potentially greater cost of delayed intervention and recovery.
Environmental Resilience — how institutions can move from recovery towards anticipation, preparedness and prevention.
Responsible AI — why explainability, uncertainty, transparency and human oversight matter when AI influences consequential decisions.
The Central Proposition
Artificial intelligence should not merely predict risk. It should help prevent harm.
Prediction without interpretation, responsibility and timely action does not create resilience.
The purpose of environmental intelligence should ultimately be to provide decision-makers with enough time, evidence and confidence to act before preventable risk becomes irreversible damage.
Beyond Wildfire
Although wildfire provides the central lens, the underlying principles extend to floods, extreme heat, drought, biodiversity loss, water insecurity, infrastructure risk, climate adaptation and disaster preparedness.
Across these challenges, decision-makers must move beyond asking only:
What is likely to happen?
They must also ask:
What should we act upon? Who should act? When should intervention occur? What is the cost of waiting?
These are questions of intelligence, judgement, governance and resilience.
From the Book to L-ARC™
The ideas explored in the book form part of a broader research agenda examining how artificial intelligence can contribute to environmental intelligence, prevention and resilience.
This work extends into L-ARC™ — the Lahiri AI–Resilience Continuum, connecting:
Anticipatory Intelligence — recognising emerging risk early enough to act.
Governance Legitimacy — ensuring decisions are explainable, accountable and trusted.
Capital Sustainability — connecting resilience action with long-term investment and institutional continuity.
Publication Details
Where to Buy
Available directly from Shroff Publishers & Distributors and through Amazon marketplaces internationally.
The proposed four-book series is intended to establish a structured body of knowledge at the intersection of artificial intelligence, environmental systems, resilience, governance and institutional decision-making.
The series progresses from foundational technical understanding to explainability, applied resilience intelligence and institutional governance. The approved titles and development statuses are recorded in the developer-ready content.
A foundational textbook establishing the data, spatial, analytical and AI concepts required to understand environmental systems and risk.
*Status: Forthcoming
An applied textbook showing how AI, machine learning and Earth observation can be used to analyse, predict and address environmental challenges.
*Status: Forthcoming
An advanced volume examining how Environmental AI can remain explainable, reliable, accountable and appropriately aligned with human decision-making.
*Status: Proposed
forward-looking volume exploring multimodal AI, foundation models, generative systems and AI agents for environmental intelligence.
*Status: Proposed
A capstone volume connecting AI-derived risk intelligence with environmental resilience, institutional decision-making, governance and implementation.
*Status: ProposedThis research examines how glacier retreat can be connected with river variability, biodiversity stress, livelihood exposure, community vulnerability and governance readiness. It applies the L-ARC™ logic of Anticipatory Intelligence, Governance Legitimacy and Capital Sustainability to the challenge of converting environmental observation into governed, funded and sustained resilience action.
Research into identifying emerging risk through the integration of infrastructure, environmental and operational variables.
Research connecting multi-hazard intelligence, governance decision-making and economic assessment.
L-ARC™ is applied retrospectively to: