All Dataleo news, jobs, analyses and tutorials around AI Agents in Supply Chain and Operations.
The role is a strong signal that AI is moving into governed product ownership for operational decisions, not remaining a collection of experiments.
The emergence of companies such as Centrum AI highlights a broader shift toward AI-powered decision layers sitting above traditional ERP and planning systems. Rather than replacing existing platforms, these solutions aim to provide risk intelligence, scenario analysis, and decision support across fragmented operational environments.
This hiring signal suggests continued investment in Supply Chain Resilience, explainable AI, and operational risk management as organizations seek better visibility into increasingly volatile global supply networks and stronger Decision Support.
This job is a strong market signal for AI in Manufacturing and Supply Chain Planning. Sanofi is not only hiring for generic digital product management; it is looking for a Product Owner able to orchestrate AI agents inside real industrial workflows, across planning, operations, quality and performance.
The most interesting element is the blend of AI Agents, industrial systems and governance. In practice, this is the profile many large manufacturers will need: someone who understands the decision architecture between ERP, MES, QMS, planning tools and AI copilots, while remaining accountable for adoption, value and compliance.
MediumMediumMediumMediumMediumHighHighHighHighHighMediumHighHighHighHighMediumKinaxis disclosed a collaboration with NVIDIA to explore long-running AI agents that could continuously optimize and adapt Supply Chain Planning decisions at scale.
HighT-Systems and SupplyOn are connecting SupplyOn’s platform to the Industrial AI Cloud to bring sovereign AI agents into industrial procurement. The first use case, AI-native Sourcing, is designed to support supplier selection, bid analysis and the preparation of sourcing decisions.
Processing will take place in a sovereign data center in Munich, reflecting European requirements around data control, security and industrial confidentiality. The partners also plan to extend the approach to planning, quality, electronic invoicing and risk management.
The announcement is an important signal that procurement agents are moving closer to enterprise workflows where recommendations can affect supplier choice, cost, risk and continuity of supply.
This is relevant for Agentic Procurement because it combines AI agents with data sovereignty and industrial governance. The central question is not only whether an agent can compare bids, but which data sources it trusts, how its evaluation logic is documented and who remains accountable for the supplier decision.
Procurement leaders should define approval thresholds, conflict resolution, audit trails and human override before agent recommendations influence awards or supplier commitments.
MediumKinaxis has published a new industry perspective on how tire manufacturers can respond to growing supply chain volatility through adaptive planning and real-time decision synchronization. The article addresses challenges including raw-material exposure, automotive OEM constraints, omnichannel demand, complex distribution networks, short lead times and reverse logistics.
The company positions Maestro and autonomous concurrent orchestration as a way to evaluate demand, supply, inventory and capacity simultaneously rather than through slow, sequential planning cycles. When conditions change, the model is intended to make the impact visible across sourcing, production and distribution so teams can respond before disruptions spread.
The tire industry is a useful example because it combines volatile natural and synthetic rubber markets, global logistics exposure, automotive service expectations, sustainability pressures and increasing product complexity from electric vehicles. Kinaxis argues that adaptive systems can help manufacturers sense changes earlier, rebalance priorities and support planners with autonomous agents rather than relying only on manual coordination.
This Kinaxis perspective is relevant for Supply Chain AI because it links agentic capabilities to a concrete decision environment. The operational value is not simply faster alerts, but the ability to evaluate service, inventory, sourcing, production and distribution consequences inside one connected decision model.
The governance question remains essential: who owns prioritization rules, substitution logic and inventory policies when an adaptive platform recommends a response? For tire manufacturers, Maestro can shorten decision latency, but companies still need trusted data, documented trade-offs, planner validation and clear boundaries between recommendation and automated execution.
HighThe U.S. Army has awarded Rune Technologies a contract valued at up to $99 million for TyrOS, an AI-powered predictive logistics platform already deployed across multiple military formations. Rune also positions Saga, its agentic logistician, as a way to compress planning cycles from days to seconds.
The announcement highlights the use of AI for resource planning, sustainment, readiness and logistics coordination in highly constrained operating environments. The platform is intended to help teams anticipate requirements and evaluate logistics options before execution.
Although the context is defense, the broader supply chain signal is the move toward AI systems that combine planning data, operational constraints and resource availability in a continuously updated logistics model.
This is relevant for Predictive Logistics because it shows agents being used where decisions are time-sensitive and operationally consequential. The same governance questions apply in industrial supply chains: data quality, traceability, scenario assumptions, human approval and fallback procedures.
The useful lesson is not the promise of instant planning, but the need to make recommendations explainable and auditable before they influence resource allocation or execution.
MediumPigment has rolled out an upgrade to its AI Agents. According to a LinkedIn post by Alexis Fromaget, Pigment’s Analyst, Modeler and Custom Agents can now pull live external context from the web during conversations and use it inside analyses, recommendations and model builds, with source citations included.
The update matters for Enterprise Planning because AI agents are moving from internal assistants toward context-aware planning collaborators. In supply chain and business planning, this can help teams connect internal models with external signals, market information, assumptions and supporting evidence.
For Supply Chain Planning, the relevant signal is not only faster analysis. It is whether external context can be used safely inside governed planning workflows, with traceability, source visibility and clear boundaries between recommendation, validation and execution.
This Pigment update is relevant for Supply Chain AI because planning agents increasingly need both internal business data and external context. The key governance question is how teams decide which external sources are trusted, how citations are reviewed, and when agent-generated recommendations are allowed to influence planning decisions.
For operations leaders, the opportunity is a more connected Decision Architecture: agents can support analysis, model building and scenario exploration, while planners retain ownership of assumptions, validation rules and final decisions. Without this control layer, live web context could add noise or unverified assumptions into critical planning models.
High
MediumColibri S&OP is positioning AI Agents inside supply chain planning workflows covering demand planning, supply planning, strategic planning, safety stock optimization and constrained plan optimization.
This is relevant for mid-market and local planning teams because it shows how agentic planning ideas are moving beyond global mega-suites. The practical question is how S&OP teams use agents to accelerate scenarios and exceptions while keeping human ownership of planning decisions.
More details are available on the Colibri S&OP website.
This product signal matters because Agentic AI in planning is not only a large-enterprise trend. Colibri S&OP should be tracked where AI helps business users structure demand, supply and scenario decisions inside a governed Planning Governance process.
HighSAP is positioning Joule Agents and Joule Assistants as context-aware AI capabilities for Supply Chain Management. The company describes these assistants as tools designed to understand business context and accelerate outcomes across logistics, manufacturing, product design, planning, and asset service workflows.
The SAP page highlights several supply chain-focused capabilities, including Logistics Assistant, Manufacturing Assistant, Product Design Assistant, Planning Assistant, and Asset & Service Assistant. This reflects SAP’s broader move to embed AI Agents directly into enterprise workflows rather than treating AI as a separate productivity layer.
More details are available on the official SAP page.
This is an important signal for Supply Chain Planning and enterprise operations teams because it confirms that the AI assistant layer is moving inside core business applications. SAP is not only promoting generic AI productivity; it is connecting Joule to operational domains where decisions depend on ERP data, process context, and business rules.
For supply chain companies, the practical question is how these agents will interact with existing planning architectures, including SAP IBP, ERP workflows, logistics systems, manufacturing execution, and asset management. The opportunity is faster analysis and better decision support; the risk is uncontrolled automation without clear AI Governance, permissions, and human-in-the-loop validation.
MediumKBRW shared that it presented how AI Agents are already creating value for large-scale supply chain operations during the Sagard NewGen 2026 Annual AGM. The signal is relevant because Sagard Europe described KBRW’s presentation as an illustration of the agentic AI shift for SaaS customers, including a deployment for a CAC 40 client. More details are available in the LinkedIn post.
For supply chain leaders, the practical relevance is the move from visibility and dashboards toward operational agents that can support exception handling, orchestration and guided action. In KBRW’s domain, this connects Order Management, Fulfillment Orchestration and Smart Steering.
HighAnthropic has released “The Founder’s Playbook,” a comprehensive guide explaining how startups can build and operate as AI-native organizations from day one. The document provides a broader view of how Generative AI and AI Agents may reshape organizational design, decision-making, and execution.
The playbook argues that AI significantly reduces the cost of experimentation and enables smaller teams to perform work that previously required larger functions. It presents AI as a research analyst, product manager, software engineer, and operational assistant working alongside human teams, while emphasizing governance, validation, and accountability.
More details are available in the official source document.
For Supply Chain Planning organizations, the playbook offers a blueprint for AI-native operating models where planners, analysts, and managers increasingly orchestrate AI-enabled workflows. Activities such as scenario analysis, forecast investigations, executive reporting, supplier intelligence, and operational monitoring could be accelerated through controlled use of Decision Intelligence capabilities.
The document also reinforces the emergence of an AI layer sitting above traditional platforms such as SAP IBP, Kinaxis, and o9 Solutions. Rather than replacing enterprise systems, AI agents can help users interpret information, generate recommendations, and shorten decision cycles while maintaining strong AI Governance and human oversight.
HighSAP has introduced its Autonomous Enterprise vision and a governed Business AI Platform. The strategy connects agents, business applications and enterprise data to automate workflows while maintaining enterprise controls.
The relevant test for supply-chain teams is whether autonomy remains bounded by decision rights, data lineage, validation and manual override across ERP and planning processes.
HighInfios has announced new AI agents embedded into supply chain execution workflows. The agents are designed to operate across orders, warehouses and transportation, supporting orchestration inside operational processes rather than sitting outside them as standalone advisory tools.
The announcement is relevant for Supply Chain Execution because it moves agentic AI closer to real operational workflows: order orchestration, warehouse issue resolution, transportation updates and exception handling. Infios positions the agents as part of its broader execution stack across OMS, WMS and TMS environments.
For logistics and operations teams, the signal is that Agentic AI is moving beyond planning and analysis into time-sensitive execution decisions. This raises practical questions around autonomy, supervision, exception thresholds and how AI-driven actions are recorded inside operational systems.
This announcement matters for Supply Chain AI because execution workflows have less tolerance for ambiguity than planning simulations. When an AI agent changes an order path, supports a warehouse supervisor or triggers a transport action, the decision has immediate operational consequences.
The key governance question is therefore not only whether the agent can act, but what it is allowed to act on. Operations leaders need clear decision boundaries, source data validation, escalation rules, audit trails and human override processes before embedding agents into Warehouse Management, Transportation Management and order execution workflows.
Highproject44 has launched Autopilot, a no-code platform for deploying and controlling supply-chain agents. The platform targets freight cost, data quality, inventory and cash-flow workflows, allowing teams to configure agent behavior without building a full custom stack.
No-code lowers the barrier to agent creation but increases the risk of shadow automation. Organizations need a registry, owner, version control and approval process for every deployed agent.
HighInfor announced an expansion of Industry AI with more than 100 agents and Agentic Orchestrator enhancements. For supply chain teams, the signal is that industry-specific enterprise applications are moving toward agentic workflows embedded inside operational processes.
For Infor Supply Chain Planning, the relevance is how AI agents could support demand planning, demand sensing, inventory and supply workflows using industry context. The governance issue is ensuring agents operate within approved planning rules and Human-in-the-Loop controls.
Infor’s Industry AI direction matters because Supply Chain AI needs industry context to be operationally useful. The practical test is whether agents improve planning responsiveness while preserving data quality, explainability and decision ownership.
HighOracle has introduced Fusion Agentic Applications for finance and supply-chain processes. The applications combine embedded agents, enterprise data and workflow context to support decisions and automate multi-step tasks.
Cross-functional agents need shared definitions and escalation rules because finance and supply-chain objectives often conflict around cash, service, inventory and cost.
Highproject44 has acquired LunaPath.ai to combine real-time logistics data with AI-agent orchestration and execution capabilities. The acquisition supports project44’s move from visibility toward autonomous response across global supply chains.
The integration challenge will be preserving data lineage, permissions and accountability as agents move from detecting events to coordinating actions across systems and partners.
Highproject44 has expanded its AI-agent portfolio across freight procurement, disruption response, carrier onboarding and related logistics workflows. The agents are built on project44’s logistics data graph and are intended to combine operational context with workflow orchestration.
The portfolio signals a broader move from point agents toward a reusable agent layer across transportation operations.
Scaling multiple agents requires shared governance: common data definitions, action permissions, conflict resolution and an auditable record of which agent influenced each operational decision.
HighHighLogility has launched DemandAI+ through Aptean AppCentral, combining AI-first forecasting with prebuilt planning agents designed for rapid activation. The offer targets demand-planning teams seeking faster deployment of automated analysis and recommendations.
HighOracle has introduced new AI agents embedded across Fusion Cloud Supply Chain and Manufacturing. The agents support planning, procurement, manufacturing, maintenance and logistics workflows, with the goal of accelerating decisions and automating repetitive operational tasks.
The move expands agentic AI across the full supply chain lifecycle rather than limiting it to a single function.
Embedded agents can shorten decision latency, but enterprises need clear ownership, access controls and validation thresholds before recommendations affect planning or execution.
HighCoupa launched new AI agents for autonomous sourcing, supplier collaboration and workflow orchestration across source-to-pay processes.
HighOracle expanded the AI agents embedded in Fusion Cloud Applications to automate Supply Chain processes, improve planning and fulfillment, and accelerate operational decisions.
Scaling embedded agents requires common governance across planning, fulfillment and logistics, with clear ownership of every automated recommendation and action.
HighGreyOrange highlighted GreyMatter as an AI-driven multiagent orchestration platform capable of coordinating large volumes of warehouse operations across robots, people and inventory.
HighInfor expanded its suite of industry-specific AI agents for operational workflows across sectors including dairy, electric vehicles and textiles. The agents are designed to work with industry context rather than generic enterprise prompts.
HighDaybreak AI raised $15 million in Series A funding from investors including TPG Growth and Dell Technologies Capital, positioning itself around AI labor for enterprise planning.
The announcement is relevant because Daybreak’s agentic framing targets repeatable planning work, policy-based automation and exception routing back to humans. This connects AI Agents, Supply Chain Planning and human-in-the-loop governance.
More details are available in the funding announcement.
This is a strong signal for Agentic AI in planning because the category is moving from copilots to AI labor for repeatable decisions. The governance question is how companies define policy boundaries, exception routing and audit trails before agents influence operational plans.
HighBlue Yonder introduced new AI agents and a Supply Chain Knowledge Graph at ICON 2025. The capabilities are designed to give agents deeper operational context across planning, logistics, fulfillment and partner networks.
HighOracle introduced AI agents embedded in Fusion Cloud Supply Chain & Manufacturing to automate end-to-end tasks and deliver role-specific insights across procurement, manufacturing, maintenance and logistics.
Embedded agents require clear permissions, validation thresholds and audit trails before they influence operational decisions.
Autonomous freight platforms are moving beyond task automation, but execution authority must be governed by lane economics, carrier commitments and compliance boundaries.
Read more →
The strategic layer for Supply Chain agents is the mechanism that controls state, authority and recovery across ERP, planning and execution systems.
Read more →
Specialised agents create value only when competing recommendations are reconciled through a common objective and escalation model.
Read more →The important distinction among monitoring, synthesis, agents, visibility and digital twins is how close each system comes to changing an operational decision.
Read more →
AI agents can accelerate application construction, but production-ready planning still depends on validated logic, data and constraints.
Read more →Operational agents become trustworthy when recommendations, approvals and commitments remain tied to evidence and current transaction state.
Read more →
The highest-value procurement decisions are often made before information enters the transactional ERP layer.
Read more →Forecast enrichment is valuable only when humans or agents contribute material information unavailable to the baseline model.
Read more →
The real test of logistics agents is their behaviour when bookings fail, data is missing or execution diverges from plan.
Read more →Mixed human and AI teams require explicit boundaries for recommendation, execution, review and escalation.
Read more →
Independent agents can optimise locally rational objectives while producing an incoherent end-to-end order decision.
Read more →
Separating AI marketing from decision capability
Read more →Lessons from Ralph Loops and goal-seeking AI for supply chain analytics and operational execution
Read more →
Lessons from Christo Delport’s analysis of AI bots, agents, and the future of supply chain intelligence
Read more →What Anthropic’s Founder’s Playbook reveals about the future of planning, decision-making, and operational work
Read more →From forecasting and procurement to governed operational decision support
Read more →From decision-ready insights to governed planning actions
Read more →Agents can investigate and recommend, but decision rights must remain explicit
Read more →