Thailand’s once-robust manufacturing sector is facing a protracted slowdown, with factory capacity utilization hovering below 60 percent for the past two years, raising concerns about the country’s economic competitiveness and industrial policy effectiveness.

Key takeaways

• Thailand’s manufacturing sector is operating at below 60% capacity for two consecutive years, with only one-third of industries recovering to pre-pandemic lockdown levels.
• Ultra-low priced Chinese imports and the influx of Chinese FDI (21% of total by 2024) are displacing Thai manufacturers, particularly in rubber, plastics, and food production sectors.
•  Stagnant credit access since 2022 is preventing Thai manufacturers from upgrading technology and innovating, trapping the economy in a low-growth equilibrium that requires long-term financial policy intervention.

The manufacturing sector, which accounts for 24 percent of Thailand’s GDP, 15.7 percent of total employment, and approximately 80 percent of exports, has been operating in the doldrums despite government stimulus measures, according to recent analysis by Professor Archanun Kohpaiboon of Thammasat University.

Pandemic Recovery Remains Elusive

Data from Thailand’s Office of Industrial Economics reveals a troubling trend: in the first ten months of 2025, only one-third of industries achieved capacity utilization rates exceeding levels seen during the strictest COVID-19 lockdown period of April-December 2021. The sectors showing resilience include beverages, leather footwear, processed foods, kitchenware, and vehicle engines.

“The low and declining capacity utilization found in many industries indicate that the demand for locally manufactured products is weak,” Kohpaiboon noted, adding that while export performance has remained stable with Thailand maintaining a 1.3-1.5 percent global market share, domestic-oriented manufacturers face particularly acute challenges.

The China Factor

Analysts point to three primary factors behind the manufacturing malaise, with Chinese economic influence looming large in each.

First, an influx of ultra-low priced Chinese imports appears to have undermined government demand-boosting initiatives. Between October 2020 and October 2023, Thailand implemented its “half-half” subsidy program five times, spending THB234.5 billion (approximately $6.5 billion) to stimulate consumer spending. However, experts suggest these programs may have inadvertently increased demand for cheap Chinese imports rather than domestically produced goods.

Second, the surge in Chinese foreign direct investment has reshaped Thailand’s industrial landscape. By 2024, Chinese investors accounted for 21 percent of total FDI inflows. While this investment has brought capital, it has also led to displacement of Thai firms in key sectors.

Between January 2021 and October 2025, 3,796 Thai manufacturing firms deregistered while 650 new Chinese firms entered the market, particularly in rubber and plastics, food production, and fabricated metal products. Many of these Chinese-owned operations maintain limited supply chain linkages within Thailand, preferring to import inputs from China and thereby reducing demand for Thai-manufactured components.

Credit Crunch Compounds Problems

The third factor is a stagnation in credit extended to the manufacturing sector. After years of steady growth, lending to manufacturers has remained virtually flat from 2022 to 2025, constraining firms’ ability to upgrade technology, pursue innovation, or explore new market opportunities.

“Businesses experienced great financial strain during the pandemic and were not able to get adequate financial support,” Kohpaiboon observed, noting that government pandemic measures focused primarily on worker relief rather than keeping businesses operational.

Call for Strategic Intervention

To revitalize the sector, experts are calling for a fundamental shift in policy approach. Rather than short-term stimulus measures, Kohpaiboon argues the government needs a comprehensive strategy to improve firms’ access to long-term financial resources.

“These activities will incur short-term investment costs and need to be carried out continuously,” he said. “They cannot be achieved by relying solely on short-term financing, such as commercial bank lending.”

The analysis warns that the current low-capacity utilization is trapping Thailand in a low-growth equilibrium, representing a critical gap in policymaking that demands urgent attention.

As Thailand navigates increasing regional competition and technological disruption, the health of its manufacturing sector will prove pivotal to the nation’s economic trajectory. With Chinese competition intensifying and domestic industrial capacity languishing, the pressure is mounting on Bangkok to craft more effective, long-term industrial policies.

Other People are Reading

Mark Dehring

Good morning, everyone. Thank you for joining CSL’s results presentation for the first half of the 2026 financial year. I’m Mark Dehring, CSL’s Head of Investor Relations. Please note, this briefing is being webcast.

We have a lot to get through today. But as usual, I’d like to draw your attention to the important disclaimer on your screen. A copy of this, along with our other ASX materials, have been published on the CSL and the ASX websites. You will also have seen announcement we made to the ASX yesterday relating to the appointment of Gordon Naylor as Interim Chief Executive Officer and Managing Director. We’ll hear from Gordon shortly.

But before we do, I’d like to introduce our other speakers today. With me here in Melbourne is Ken Lim. Ken has been our CFO since October last year. Prior to that, he was our Chief Strategy Officer and has previously led our Seqirus business unit. Also here is Chief Commercial Officer, Andy Schmeltz. Andy joined CSL in 2023 as Executive Vice President, CSL Behring. And last year, his role was expanded to include CSL Vifor. And finally, we have Dave

As we move deeper into the second half of the decade, businesses across every industry are recalibrating their digital strategies around artificial intelligence.

Whether you’re a startup founder, CTO, or part of an AI software development company, the accelerating pace of innovation is reshaping how systems are built, deployed, and maintained. 2026 is proving to be the most transformative year yet, with breakthroughs not only in model capabilities but also in the frameworks, ethics, infrastructure, and methodologies that support them. Below are the defining trends that are shaping this rapidly evolving landscape—and why they matter for the next generation of intelligent software.

Top 10 AI Trends to Watch in 2026

As artificial intelligence matures at an unprecedented rate, 2026 is emerging as a pivotal year for software development innovation. From agentic systems that can autonomously build and optimize applications to multimodal models capable of understanding the world across text, visuals, and sound, AI is reshaping how digital products are conceived and delivered. These trends are not just influencing technical workflows—they’re redefining business strategy, product design, security standards, and user experience across every industry. Below are the ten most significant AI shifts shaping the future of software development and what they mean for organizations preparing for the next era of intelligent systems.

1. The Rise of Agentic AI Systems

Preview: AI shifts from passive assistants to fully autonomous agents capable of handling complex tasks, analyzing workflows, and self-correcting without human intervention.

In 2026, AI is moving toward agentic systems that can independently execute tasks, collaborate with other agents, and optimize their own performance. These agents are capable of generating production-quality code, debugging applications, orchestrating cloud resources, and monitoring system behavior. This transformation is redefining software development by reducing manual involvement and amplifying engineering productivity.

2. Multimodal AI Becomes a Standard Development Tool

Preview: Models that understand text, images, audio, video, and sensor data simultaneously become foundational to modern apps.

Multimodal AI—once considered experimental—has become essential. Developers now build systems that interpret real-time video, analyze medical imaging, manage robotics, and generate highly creative visual and auditory content. By merging modalities, AI gains near-human perception, unlocking new technical and creative applications across industries.

3. AI-Native Application Architecture

Preview: Software is increasingly designed from the ground up with AI at its core, not as an add-on feature.

Just as cloud-native design transformed development a decade ago, AI-native architecture is now the new gold standard. These applications incorporate continuous learning, real-time inference pipelines, multi-model orchestration, and rigorous model lifecycle management. AI becomes the heart of the system, enabling applications that evolve alongside business needs.

4. The Maturation of Synthetic Data Pipelines

Preview: Companies turn to fully developed synthetic data ecosystems to overcome privacy, scarcity, and cost challenges.

Synthetic data has become a necessity rather than an optional enhancement. In 2026, hyperrealistic simulations power robotics and autonomous systems, while synthetic tabular data supports finance, healthcare, and government AI. AI-to-AI data generation accelerates training, lowers risk, and boosts accuracy—especially in domains where data collection is limited or regulated.

5. Privacy-Preserving AI and Secure Model Development

Preview: Stricter regulations push companies to adopt secure AI practices like federated learning, encrypted computation, and differential privacy.

AI governance laws worldwide are compelling teams to rethink how data and models are handled. Techniques like encrypted computation (FHE), zero-knowledge proofs, federated learning, and differential privacy have become integral to modern AI development. This ensures that models remain powerful while meeting global compliance standards.

6. The Expansion of Low-Code / No-Code AI Development

Preview: AI-augmented platforms enable non-engineers to build functional applications using natural language and visual tools.

Low-code platforms have evolved dramatically thanks to agentic AI. Anyone can generate apps through natural language, connect datasets, automate workflows, and deploy AI services without writing extensive code. While traditional development remains essential, low-code dramatically accelerates prototyping and empowers business teams to innovate independently.

7. Micro-Models and Domain-Specific AI

Preview: Specialized lightweight models replace one-size-fits-all giants, enabling faster, cheaper, and more accurate task performance.

Micro-models are optimized for specific industries or tasks—legal work, materials science, edge computing, and embedded robotics. They run faster, require fewer resources, and deliver higher accuracy within their domains. This shift toward modularity is making AI more scalable, efficient, and industry-specific.

8. Generative UI and Adaptive User Experience

Preview: Software interfaces become dynamic, adjusting in real time based on user behavior, experience level, and preferences.

Generative UI allows applications to rewrite their own interfaces based on user interactions. Dashboards rearrange automatically, workflows adapt to user proficiency, and customized visualizations are generated on demand. This creates ultra-personalized user experiences that enhance productivity and reduce friction.

9. AI-Optimized DevOps and Autonomous CI/CD

Preview: DevOps evolves into a largely automated ecosystem where AI predicts issues, resolves failures, and optimizes deployments.

In 2026, AI-driven DevOps systems detect integration risks before they occur, identify root causes instantly, automate code rollbacks, optimize cloud spending, and manage deployment pipelines without manual oversight. These self-healing systems reduce downtime dramatically and free developers to focus on high-impact work.

10. Ethical, Transparent, and Responsible AI Becomes Non-Negotiable

Preview: Regulations and public expectations require detailed transparency, explainability, bias detection, and auditability in AI systems.

As AI powers critical systems globally, ethical development is now mandatory. Responsible AI frameworks ensure transparency, explainability, fairness, and accountability. Companies that embed responsible practices gain trust, avoid legal consequences, and ensure long-term sustainability of their AI strategies.

Looking Ahead

2026 marks a defining moment for AI-driven software development. With agentic AI, multimodal intelligence, synthetic data, privacy-preserving methods, and autonomous DevOps, the future of software is adaptive, self-evolving, and deeply integrated with intelligent systems. Organizations that embrace these trends early will lead innovation, deliver superior products, and achieve a competitive advantage in a rapidly shifting digital world.

FUNDAMENTALS

Spot gold edged 0.3% higher to $5,038.73 per ounce by 0059 GMT.

U.S. gold futures for April delivery gained 0.6% to $5,060.60 ‌per ounce.

⁠Spot ⁠silver was up 1% at $81.49/oz, after falling more than 3% in the previous session.

Live Events

U.S. yields fell on Tuesday after a raft of data suggested the economy may be softening, giving the U.S. Federal Reserve more cushion to cut interest rates. [US/]
Falling yields reduce the cost of holding metals and ​often come with macro signals that favour ⁠them.U.S. ‌retail sales were unexpectedly unchanged in December as ​households scaled ​back spending on motor vehicles and other big-ticket items, ⁠potentially setting consumer spending and the economy on a ​slower growth path heading into the new year.

Federal Reserve Bank of Cleveland President Beth Hammack, however, said on Tuesday that the U.S. central bank faces no urgency to change the setting of interest rates this year amid a “cautiously optimistic” outlook for economic activity.

Investors expect at least two 25-basis-point rate cuts ‌in 2026, with the first one expected in June. Non-yielding bullion tends to do well in low-interest-rate environments. [FEDWATCH]

Investors ​await the ​non-farm payrolls report ⁠for January, due later in the day, and inflation data on Friday for more cues on the Fed’s monetary policy path.

Indian investors piled ​into gold exchange-traded funds in January as prices soared amid rising geopolitical risks, surpassing flows into equity funds for the first time, industry data showed on Tuesday.

Spot platinum added 0.6% to $2,098.78 per ounce, while palladium rose 0.2% to $1,712.25.

DATA/EVENTS (GMT)

0130 China PPI, CPI YY January

1330 US Non-Farm Payrolls January

1330 US Unemployment Rate January

1330 US Average Earnings YY January

Market Overview

Global fixed-income markets delivered positive total returns in the fourth quarter of 2025 as policy uncertainty, fiscal developments, and divergent central-bank actions shaped the investment landscape. The US began the quarter with its longest government shutdown

Operator

Good day. Welcome to Upexi Inc. Fiscal Second Quarter 2026 Financial Results Conference Call. Please note this event is being recorded.

I would now like to turn the conference over to Valter Pinto, Managing Director at KCSA Strategic Communications. Please go ahead.

Valter Pinto
Kanan, Corbin, Schupak & Aronow, Inc.

Thank you, operator. Good evening, and welcome, everyone, to the Upexi Fiscal Second Quarter 2026 Financial Results Conference Call. I’m joined today by Allan Marshall, Chief Executive Officer; Andrew Norstrud, Chief Financial Officer; and Brian Rudick, Chief Strategy Officer.

Before we begin, I’m going to remind everyone that statements made during today’s conference call may be deemed forward-looking statements within the meaning of the safe harbor of the Private Securities Litigation Reform Act of 1995. Actual results may differ materially due to a variety of risks, uncertainties and other factors. For a detailed discussion of some of the ongoing risks and uncertainties in the company’s business, I’ll refer you to the press release issued this evening and filed with the SEC on Form 8-K as well as the company’s reports filed periodically with the SEC. The company disclaims any intention or obligation to update or revise any forward-looking statements, whether as a result of new information, future events or otherwise, unless otherwise required by law.

In addition, during the course of the call, we may refer to non-GAAP financial measures that are not prepared in accordance

Jaguar Land Rover is recalling nearly 2,300 electric SUVs in the U.S. over concerns that a high-voltage battery may overheat, increasing the risk of fire, the National Highway Traffic Safety Administration (NHTSA) announced on Tuesday.

The recall impacts 2,278 Jaguar I-Pace vehicles from the 2020–2021 model years.

“As an interim repair, the battery software will be updated by a dealer, or through an over-the-air (OTA) update to limit the state of charge to 90%” the NHTSA said, according to Reuters, adding that the final remedy is currently under development.

BMW RECALLS NEARLY 90,000 VEHICLES OVER ENGINE STARTER FIRE RISK

Jaguar Land Rover is recalling nearly 2,300 electric SUVs in the U.S. over concerns that a high voltage battery may overheat, increasing the risk of a fire. (Sjoerd van der Wal/Getty Images / Getty Images)

There will be no charge to vehicle owners for the interim repair, the recall report said.

Customers can monitor their vehicle’s charging with the latest version of the Jaguar Remote App or inside the vehicle, according to the report, which says owners should physically stop charging by unplugging the cable when it reaches a 90% state of charge.

Vehicle owners are urged to park outside and away from structures and to charge outside if possible.

The recall impacts 2,278 Jaguar I-Pace vehicles of the 2020-2021 model year. (Aly Song/Reuters / Reuters Photos)

“Vehicles have experienced thermal overload, which may show as smoke or fire, that may occur in the high voltage traction battery pack. The investigation is ongoing,” the report reads.

Investigations pointed to a “folded anode tab” in battery cells produced at an LG Energy Solution facility in Poland, which can lead to short-circuiting.

CHRYSLER RECALLS MORE THAN 450,000 VEHICLES OVER BRAKE LIGHT FAILURE

Investigations pointed to a “folded anode tab” in battery cells produced at an LG Energy Solution facility in Poland. (Anna Barclay/Getty Images / Getty Images)

GET FOX BUSINESS ON THE GO BY CLICKING HERE

“Modules that were identified by the remedy software as having characteristics of a folded anode tab, which may contribute to a risk of thermal overload, are still being inspected by the supplier,” it added.

Notification letters are expected to be mailed to affected owners starting April 3.

A top official involved in the Trump administration’s new AI “Tech Force” initiative offered a glimpse inside the White House’s push to place AI engineers inside federal agencies and overhaul how the federal government uses technology.

“This is a public-private partnership where we’re going to bring a thousand great engineers into government to help us complete the modernization that is so important to our American people and so important to making sure that technology is a first-class citizen inside of government,” Scott Kupor, Office of Personnel Management director, said Tuesday on “Varney & Co.“

With help from companies such as Microsoft, Palantir, Salesforce and Snowflake, Kupor said the initiative is designed to embed engineers across major federal agencies rather than centralizing them in a single office, allowing technologists to work directly on modernization efforts at departments including the Treasury Department, the Department of Health and Human Services and the Department of War.

TRUMP ADMIN REUNITES WITH ELON MUSK IN PURSUIT OF AI DOMINANCE: ‘BENEFIT OF THE COUNTRY’

President Donald Trump speaks in the Oval Office at the White House on October 06, 2025 in Washington, DC.  (Anna Moneymaker/Getty Images / Getty Images)

Palantir, for one, already plays a key role in integrating artificial intelligence into the Department of War, Kupor noted, pointing to the company as an example of how advanced technology can be deployed effectively inside government.

“Palantir is no doubt a leader in that area, and what we want to do is make sure that other parts of technology, other parts of AI get into government,” he explained. 

CHEVRON CEO DETAILS STRATEGY TO SHIELD CONSUMERS FROM SOARING AI POWER COSTS

Palantir is one of the major tech companies behind the Trump administration’s AI push. (Rafael Henrique/SOPA Images/LightRocket via Getty Images / Getty Images)

“Today, what happens is we have a very heavy reliance on contractors and consultants, as you may know. We don’t have as much homegrown talent in core AI, modern software development, and Tech Force is really intended to make sure that the government can be a first-class citizen and attract those types of individuals, just like Palantir has done in the government, but we also want to make sure that talent is resident in government long-term.”

Kupor said the engineers brought in through Tech Force will be placed directly inside agencies and supported through a structured program that includes professional development, a speaker series and engagement with private-sector partners.

He said applications for the initiative are already coming in, with the administration expecting to begin making offers within the next 30–45 days.

GET FOX BUSINESS ON THE GO BY CLICKING HERE

At the end of the program, many workers are expected to move into private sector roles, taking with them experience gained working inside the federal government.

“The whole goal is, agency by agency, [to] figure out what are all the major modernization efforts and how do we make technology, how do we make AI really important to the American people?” he said.