🔦 Track 6: Simple Agentic Use Cases for Real-World Operations

Saudi Arabia is rapidly scaling smart infrastructure, autonomous systems, and data-driven operations as part of Vision 2030.Across mobility, industry, and infrastructure, machines and assets are becoming increasingly connected — yet many operational processes still rely on manual decisions, approvals, and payments.

While AI systems already generate insights, the next step is controlled action:AI agents that can act on behalf of machines or assets, safely, transparently, and within clearly defined limits.

The challenge is not technological possibility — it is trust, accountability, and business viability.

📌 Relevant Context & Inspiration (Selected References)

• Saudi Vision 2030 – Smart Infrastructure & Digital Transformationhttps://www.vision2030.gov.sa/
• Autonomous & smart mobility initiatives in Saudi Arabiahttps://www.misa.gov.sa/
• Industrial digitalization & smart manufacturing programshttps://www.modon.gov.sa/
• Agentic AI & autonomous decision-making in enterprise systems(General industry research and applied enterprise use cases)
• Digital identity, auditability & secure automation (Web3 concepts applied to enterprise workflows)(Decentralized identity, verifiable credentials, smart contracts)
• Fatch.AI https://www.fetch.ai
• As a example: DID eIDAS2.0 https://commission.europa.eu/topics/digital-economy-and-society/european-digital-identity_en

📝 Note:The references above are only a small selection to help teams and mentors understand the broader context and relevance of this challenge.

Teams are not expected to replicate existing solutions or follow predefined architectures.

What This Spotlight Is Encouraging

We strongly encourage simple, focused, and realistic use cases.

Ideas do not need to be complex, futuristic, or fully autonomous.

Instead, teams should focus on:

• One clear operational problem
• One specific asset or machine
• One agent performing one action
• One clear beneficiary and payer

Ideas May (and Should) Also Include

• Clear business models(Who pays? Why? Per asset, per action, subscription, SLA-based?)
• Practical go-to-market strategies(Pilots, enterprise onboarding, regulatory alignment)
• Adoption models for operators and non-technical users
• Partnerships with:Asset ownersFleet operatorsInfrastructure providersFintechs or service providers

What This Challenge Is Really About

This challenge is about reimagining how machines participate in economic and operational workflows — not as uncontrolled autonomous systems, but as trusted, limited, and accountable actors.

From concept to real-world deployment, the focus is on:

Simple use cases that could realistically be approved, piloted, and paid for.

🔧 Bosch Perspective: Responsible Autonomy in the Real World

As a global industrial and mobility company, Bosch operates in environments where AI systems directly interact with the physical world — vehicles, machines, infrastructure, and safety-critical processes.

In these environments, autonomy must be controlled, understandable, and trustworthy.AI is not evaluated by its theoretical potential, but by how reliably it performs clearly defined actions under real-world constraints.

From Bosch’s perspective, meaningful agentic systems:

• Act on behalf of a clearly defined asset or system
• Operate within strict permissions and boundaries
• Are transparent and auditable
• Improve real operational processes rather than adding complexity

The focus is not on building fully autonomous systems, but on enabling responsible autonomy:

AI agents that take limited, well-understood actions — while humans remain in control.

This is why simple, precise use cases are essential.Only solutions that are easy to explain, easy to trust, and easy to integrate can realistically transition into real industrial and mobility environments.