Small company disrupts traditional industries with AI: logistics giant's market value evaporates by 23.3 billion yuan, disruptor's stock triples in two days. Beware of "information matching" business.

AI (Artificial Intelligence) Disrupts Traditional Industries — Real Cases with Astonishing Impact

On February 12th, Eastern Time, an unnoticed news broke: AI logistics company Algorhythm Holdings [RIME.O] (hereafter Algorhythm) released an industry white paper announcing that its logistics platform SemiCab, driven by AI optimization and highly profitable SaaS (cloud-based subscription software), automates processes to reduce freight empty miles while maintaining low operating costs, resulting in a threefold increase in productivity.

Once the news was released, the capital markets seemed to sense a crisis. The US logistics sector stocks plummeted sharply. The Russell 3000 Road Transportation Index temporarily dropped over 9% intraday, closing down 6.6% that day — the largest single-day decline since April 2025, when Trump’s tariff policies were announced. Light-asset logistics giant Robinson Global Logistics’ stock fell nearly 15%, reducing its market cap by about 23.3 billion yuan, with a 24% intraday plunge — the largest in history. Freight matching service provider Ledi Transportation’s stock declined 16%.

In contrast, Algorhythm’s stock rose against the trend, closing nearly 30% higher on February 12, and on February 13, surged by 222.22%, from $1.08 to $3.48. Within two days, its market value tripled.

Robinson Global Logistics is a leading global light-asset logistics company, owning no trucks, ships, or aircraft. It integrates 450,000 contracted carriers to provide diversified logistics services to 83,000 clients.

Algorhythm fully transitioned to AI logistics in 2024 and acquired SemiCab in 2025. Based on the last trading day before its market cap sharply fluctuated — February 11 — its valuation was approximately $33 million. Compared to Robinson’s $160 billion and Ledi’s $38.3 billion, it’s a tiny player in the sector. Yet, this seemingly insignificant company used AI to break through the “moat” of traditional logistics overnight.

AI Triples Freight Platform Labor Productivity

Why is an AI-driven automation software so devastatingly powerful?

Research shows that in markets like India and the US, 30% to 35% of truck miles are empty, due to dispersed planning leading to underutilized assets.

On February 12, 2026, Algorhythm released its white paper, stating that its AI-based cloud collaboration transportation platform SemiCab can expand freight volume by 300% to 400% in actual deployment. Some operators using SemiCab manage over 2,000 shipments annually without needing additional staff, compared to the industry standard of about 500 per freight broker. This indicates a threefold increase in labor productivity.

The white paper explains that in highly fragmented markets, integrating shippers, routes, and regional demand and supply can reveal return routes and cross-route flows invisible at the contract level. For example, in India, implementing this model has shown the potential to reduce empty miles from 30-35% to below 10%, without renegotiating contracts or changing carrier behaviors.

According to its official website, AI uncovers efficiencies that traditional freight management systems cannot. Through a scalable global SaaS platform, automating workflows reduces manual planning, accelerates load execution, automatically identifies optimal load combinations, reduces empty miles, and enhances network profitability.

Algorhythm claims that traditional transportation management systems and broker platforms rely on manual optimization layered on static rules. While effective at low to medium capacities, their efficiency diminishes as complexity increases.

Despite its quarterly revenue ending September 30, 2025, being less than $2 million and net losses near $2 million, its stock surged 82% after the announcement, eventually closing up 30% at $1.08, and further soared to $3.48 on February 13, 2026.

Algorhythm believes that AI-enabled operational leverage will become a key feature of next-generation logistics networks.

On February 13, 2026, Robinson Global Logistics also noted in its annual report that competitors are leveraging advanced digital platforms, AI-driven freight matching, and automation to improve efficiency and reduce costs. If the company cannot keep pace in automation and AI adoption, it may fail to achieve strategic goals in operational efficiency and digital transformation.

The “Pre-made Meal” Era in Software Industry?

SemiCab’s platform could indeed solve pain points in freight, potentially making traditional transportation management systems obsolete. However, the issue of empty miles has long plagued the freight market. Why is a platform capable of disrupting freight industry only now emerging, driven by AI?

To explore this, the Daily Economic News interviewed several AI experts.

Question 1: SemiCab runs on AI. At which stage of software development can AI play a role? How does it differ from traditional software development?

Technology investor and Unknowable AI Research Institute Director Du Yu:

Traditional development is like “building an entire building from scratch by hand.” AI-based software development is more like “having standardized structures and pipelines first, then customizing and rapidly renovating with AI and humans,” and AI can directly “check structures, find interfaces, tune interfaces” using tools, reducing time spent on reviewing documents, matching fields, and writing glue code. It’s like the “pre-made meal era” in software.

Question 2: Why wasn’t this tool developed by traditional big software firms? Is it because traditional software capabilities are insufficient, or does AI’s foundation bring advantages that traditional logic cannot?

China Ministry of Industry and Information Technology Expert Committee Member Pan Helin:

Improving the utilization rate of return empty miles can greatly boost logistics efficiency. Many companies have tried this, but the persistent problem is that freight information is highly fragmented. Users post info on different platforms; LTL freight sometimes needs to be combined to meet a shipment’s needs.

Integrating these scattered freight orders is a tough challenge. I believe SemiCab, as an aggregation platform, can solve part of the empty load problem but isn’t a definitive solution with high potential. However, AI aggregating information is a promising approach.

Question 3: Why has the freight empty mile problem persisted for years, yet only now can an AI platform truly solve it?

Liu Gaocang, Executive Deputy Director of Guojin Securities, leading tech initiatives:

Empty miles aren’t just a “problem of someone thinking about it,” but a systemic challenge. Traditional software development paradigms can’t handle this complexity.

In traditional models, freight platforms rely on rule engines, experience, and local optimization, dealing with highly fragmented demands (time, routes, vehicle types, shipper credit), constantly changing states (cancellations, price fluctuations), and long decision chains with many exceptions. These systems aren’t “illogical,” but “state spaces are too large,” making rules increasingly complex and diminishing returns, ultimately requiring more manpower.

The new AI platform represented by SemiCab is a victory of innovation and a fundamental shift in “development paradigm.”

It signifies a move from “rule-based” to “probability and prediction-driven” approaches: traditional freight software is built on rigid rules (If-Then). Facing massive, dynamic, fragmented orders and routes, traditional algorithms struggle to perform optimal global dynamic matching quickly. AI provides the capacity to process high-dimensional data.

Furthermore, software development faces a “cost reduction and efficiency enhancement” critical point: building such complex matching systems used to be costly and difficult. AI provides generalization, enabling software to “understand” business flows without writing redundant code for every special case.

Liu believes that traditional software can support “tools,” but only AI provides the “brain.” The advantage of AI is breaking the linear “more manpower equals more output” limit, enabling exponential productivity gains.

Question 4: Does the release of this software indicate a paradigm shift in software development? What impact does AI have on the industry?

Du Yu:

The logic is indeed changing, shifting from “coding to deliver features” to “using standard bases + AI to turn delivery units into ‘processes’.” Previously, software companies resembled “handicraft workshops,” doing one project at a time; now, they are more like “central kitchens for pre-made dishes + AI chefs”—standardized ingredients (general capabilities) mass-produced, with AI quickly assembling the “dish” (business process), and developers focusing on “cooking” (business correctness, performance, security, maintainability).

This “standard base + AI for rapid customization” approach has strong market potential domestically, but competition will be fierce: whoever can turn industry know-how (how to do) into repeatable modules, data, and process templates will succeed. Simply “knowing how to code” will become less valuable, while “industry understanding + implementation + continuous iteration” will be more valuable.

What AI can disrupt and what it cannot

Question 5: If software development becomes easier, will newly developed software be easily replicated? Will profit models based on software development be overturned? If software development ability is no longer a barrier, how can companies maintain their uniqueness?

Du Yu:

It will be easier to replicate “superficial functions,” but much harder to replicate “effective systems.”

AI makes “writing code” cheaper, but “running reliably, long-term, and winning” becomes more scarce. Functionality can be copied, but systemic capabilities and organizational competence are hard to duplicate. Profits relying on “software development” will be squeezed but not eliminated—rather, they will diversify: pure outsourcing, time-and-material billing, headcount stacking will see continued profit compression due to AI.

As software increasingly resembles “commodities,” clients care more about “who is responsible if something goes wrong” and “long-term support.” These are business relationships and responsibilities AI cannot replace. AI lowers the barrier to “making software,” but raises the barrier to “making software a business, a system, a standard.” China’s market is expected to amplify this effect.

Question 6: Which industries might be fully disrupted (completely replaced or needing a fundamental change to survive), and which will not?

Pan Helin:

Industries vulnerable to AI disruption include: one, information matching sectors like search, shopping, freight — fundamentally about connecting information; two, content creation fields like video, images, text, code.

Liu Gaocang:

I prefer to classify based on “whether a fundamental production mode overhaul is necessary,” rather than “will be replaced.” This includes:

Most vulnerable to AI disruption:

1. Repetitive, intensive operations: e.g., traditional logistics operations handling 500+ trips annually.

2. High information density but low action cost: AI decision-making can be directly executed, e.g., internet services, back-office finance.

3. Highly standardized workflows with quantifiable results and feedback, e.g., software development.

Less vulnerable:

1. Industries relying heavily on complex real-world interactions with high execution costs, such as offline services, complex manufacturing, certain medical scenarios. AI acts more as an “augmenting tool” than a replacement. For example, AI can dispatch trucks, but roadside mechanics or complex accident responders require physical dexterity and on-site adaptability that robots can’t fully replicate soon.

2. Fields involving authority and responsibility judgments, with strict regulation and low tolerance for errors: legal decisions, high-level medical judgments, major investment decisions. AI can provide data support, but humans must retain the final responsibility.

Overall, AI isn’t about “eliminating industries,” but about forcing some to upgrade their production modes. Those that don’t will be eliminated, but the industries themselves will continue in new forms.

Question 7: From this perspective, if the steam engine was an evolution in energy use, productivity, and production methods, how has AI sharpened human senses and driven evolution in which areas?

Pan Helin:

Currently, AI’s main application is improving information acquisition efficiency. For example, AI in shopping enhances transaction matching, connecting supply and demand faster. Similarly, AI freight improves efficiency. In the AI era, information supply and demand are matched more quickly than in the internet era, raising societal efficiency. If the internet era’s problem was information overload, AI enables precise matching of information supply and demand.

Liu Gaocang:

The steam engine freed humans from physical labor, and AI has driven evolution in three aspects:

1. Perception dimension: AI can process multi-source information inputs far beyond human limits (e.g., orders, prices, routes, real-time statuses in freight markets), giving organizations “global perception” for the first time.

2. Production paradigm: The cost of accessing knowledge drops sharply. Human evolution shifts from memorizing knowledge and learning skills to defining problems.

3. Organizational form: Systems like SemiCab essentially abstract “industry experience” from individuals into repeatable software capabilities, exponentially expanding management scope and organizational leverage.

In this sense, AI doesn’t just improve single-point efficiency but clarifies which complex problems should be softwareized. Humans will focus more on goal setting and value judgments, while many intermediate layers will be reconstructed by AI.