TECH & AI learning for processing sensor data. In addition, we use this formal logic as a safety harness around decisions from generative and world models. These models can still provide creative solutions to unforeseen circumstances, but they must be strictly validated for safety before execution. Finally, we use LLMs for human-machine communication. We call this hybrid AI framework, pioneered by my team at Hitachi, Reliable AI.
Premkumar: Safe industrial AI is achieved through engineering rigour and architectural grounding, not probabilistic optimism. First, AI must be designed into the system, aligned with physics, control theory and operational constraints – not bolted on as an opaque model. This is where Neurosymbolic AI becomes critical. By combining data driven learning with symbolic reasoning, rules and domain knowledge, we can ensure models are grounded in first principles, operating envelopes and known system behaviour. Second, we apply formal validation and verification techniques – using symbolic constraints to prevent unsafe actions, enforce invariants and operating limits and validate model behavior before and during deployment.
Third, AI must operate within humanin-the-loop and human-on-the-loop architectures, where accountability and authority remain clear. Finally, this is reinforced by continuous monitoring, explainability and lifecycle governance, so models remain reliable as systems,
“ AI will evolve from a passive advisory layer into an active execution layer”
Yuriy Yuzifovich Chief Technology Officer of AI at GlobalLogic, Hitachi Group Company Hitachi
data and conditions evolve. In industrial environments, learning alone is not enough. Trust comes from AI that can reason, explain and prove its behaviour – especially when the cost of failure is measured in safety, uptime and societal impact.
Ram: Realising AI safety necessitates a rigorous, engineering-led approach which in turn requires meticulous model testing, transparent procurement, full auditability and secure deployment capabilities. Alongside this, human-in-the-loop design and a holistic focus on functional safety and cyber resilience is needed, designing for security from the ground up to protect critical infrastructure.
manufacturingdigital. com 73