Since ChatGPT’s launch, generative AI has seeped into every corner of society. Now, agentic AI is the latest buzzword, but what are these tools? And how do they make sense for logistics?
Big shippers and TMS providers have been rolling out AI in recent years to save money and make their systems easier to use. Yet by the end of 2024, 42% of companies abandoned their pilot projects, up from 17% the previous year.
However, recent research released by McKinsey found that the investments have, in fact, paid off. The report notes that autonomous routing and scheduling led to over 20% cuts in inventory and logistics costs.
Earlier insights from Microsoft mirror this, finding that these tools helped slash expenses by 15%, lift service levels by 65%, and could propel the industry toward $1.3–$2 trillion in annual economic value.
So what’s different between last year’s findings and more recent results?
Let’s look at how smarter algorithms and AI agents that act, plan, and adapt in real time, change logistics. For good or bad.
Read also | Our story on How Artificial Intelligence Is No Longer an Option in Logistics
What’s happening now
The global generative AI in logistics market size was valued at $1.3 billion in 2024 and is projected to grow at a CAGR of 33.7% between 2025 and 2034. Reasons for its growth are the personalized service it helps logistics companies offer.
Generative AI is built on large language models that can process and write content in the human language, like shipping documents or email responses, based on the data you give it. Unlike traditional analytical tools, it can complement its analysis with summaries and explanations to help leaders digest key information more quickly and in a language they speak. While you don’t need to be an expert to understand its responses, you do need to be knowledgeable in the field to spot a recommendation that sounds off.
Since these tools process huge datasets, they can hiccup in certain areas. For example, a generative AI tool might misinterpret an unusual customs code or suggest a delivery time that ignores local driving restrictions. This is why it’s important for operators to build use cases with well-thought-out (or fairly straightforward), process-oriented workflows, and review AI-generated suggestions, applying their own industry knowledge before acting. Here are some successful use cases we are seeing today:
- AI agents control freight spending and reduce their carbon footprint.
- They scan shipping documents, verify information, and send them where they need to go.
- $85 million is being raised for logistics, quoting, dispatch, tracking, appointment scheduling, document collection, and billing agents.
The largest proportion of leaders who have applied generative AI (30%) say that their organization checks up to 20% of gen-AI-produced content before use. However, tracking well-defined KPIs for gen AI solutions has the most impact on the bottom line, while establishing a clearly defined road map to drive adoption of gen AI also has one of the biggest impacts at larger organizations.
Read also | Our story on How GenAI Agents Are Transforming Industry
Why it matters to logistics leaders
Generative AI sets the ground for collaborative systems that analyze and generate content at different stages of a completed workflow, together creating autonomous supply chain ecosystems. The concept of agentic AI is to act on content, automating decisions and tasks across the logistics chain, leading to self-sufficient operations. In other words, entire market business models will change.
Stock Titan found that enterprises failing to effectively implement AI risk losing 8.6% of monthly revenue, averaging $87 million annually per company. However, Boston Consulting Group estimates that only 26% of companies have developed the necessary capabilities to move beyond proofs of concept and generate tangible value from AI. Everyone is ready to start their journey; it’s just about starting small with one pilot project at a time.
We’re at that sweet spot where enough early adopters have made experiments, identified the tools’ flaws, and printed playbooks on how to overcome them. But the products are still new enough to give the next sweep of buyers a competitive edge. Best practices include redesigning workflows and adding generative AI oversight to senior leaders’ priorities and job descriptions.
If we consider the AI agent freight planning assistant, you can start by automating repetitive logistics tasks like auditing freight invoices, coordinating with carriers, or tracking shipment delays. These tools can ingest data from Excel files, but the more structured and complete your data, the better they can automate and optimize your workflows. Typical preparation involves identifying the main processes you want to automate and organizing your key logistics data (even if it’s in spreadsheets).
Read also | Will GenAI Companies Ever Make Money?
What’s next with AI
When I proposed earlier that entire business models would change and that self-sufficient operations would be possible, I meant it. But for these to work smoothly, human operators must drive each stage, verifying and correcting the tools.
Looking forward five or so years, we can start to imagine predictive models orchestrating most stock levels, delivery schedules, and vehicle routing. This is already the case for Amazon and DHL. The human layer will validate expectations, such as sudden weather disruptions, regulatory changes, or customer escalations, and handle strategic decisions like capacity reallocation.
Fast forward another ten years, and some retailers may begin outsourcing their “last-mile” and “near-mile” logistics to a network of AI-powered hubs, reducing delivery times and cutting emissions. This could lead to subscription-based business models for businesses to access autonomous micro-hub networks without needing to invest in infrastructure themselves.
For the foreseeable future, we’ll still need drivers in trucks and human planners monitoring analytics, but the user interfaces will be more accurate and more intuitive. Features such as assisted steering and braking will be widely trusted, and drivers will continue to familiarize themselves with advanced routing tools and communication platforms to keep stakeholders updated.
Tools like Tacho 2, which are mandatory since August 2025, will reduce the need for physical document border checks, allowing for smoother, faster journeys. Advanced telematics will help keep back office teams informed about how vehicles are performing, which routes are fuel-efficient, and plan better schedules.
AI advances are slowly reshaping touchpoints across the entire supply chain, giving human planners the data they need in visual and written formats that make sense. With more data-driven processes, fleet operators can streamline operations and cut carbon footprint; this is the big picture. Each individual process we automate and streamline today will help us get there.
Read also | Who Leads the Chatbot Revolution?
