Coercion hardens Taiwan’s views; education diplomacy builds trust Extend China’s regional magnanimity to Taiwan with visas, credit recognition, scholarships, and joint labs Protect academic freedom, track mobility and co-production, and scale successes to shift opinion
Adjusted deflators reveal hidden gains in US manufacturing productivity The bottleneck is skills, not machines Scale capital-intensive training, stackable credentials, and adult apprenticeships
The standard view sugges
Services trade is increasingly borderless and digital Tariffs miss; data rules, licensing, and standards decide access Train for exportable skills and build trust-based regimes to unlock growth
In 2024, the world exporte
General AI predicts probabilities, not context-specific safety Domain-specific AI fits the task and lowers risk in classrooms and markets Use ISO 42001, NIST RMF, and the EU AI Act, and test on domain benchmarks
Reported AI incidents reached 233 in 2024, marking a 56% increase from the previous year. This sharp rise highlights real-world failures as generative systems become part of everyday tools. The public is aware of this shift. By September 2025, half of U.S. adults expressed more concern than excitement about AI in their daily lives, with only 10% feeling more excited than worried. This sentiment reveals a trend that cannot be ignored: general-purpose models focus on predicting the next token instead of meeting safety standards that differ by context. They operate on probability, not purpose. As usage shifts from chat interfaces to classrooms and financial settings, relying on probability as a measure of safety is inadequate. The solution lies in domain-specific AI, which consists of systems designed specifically for a narrow task, including safety measures suited to the task's stakes.
The main issue with the push for greater generality is that it misaligns model learning with safety requirements. Models learn from training data trends, while safety is situational. What is acceptable in an open forum may be harmful in a school feedback tool or a trading assistant. As regulators outline risk tiers, this misalignment becomes both a compliance and ethical concern. The European Union’s AI Act, which took effect on August 1, 2024, categorizes AI used for educational decisions as high-risk, imposing stricter obligations and documentation requirements. General-purpose models also fall under a specific category with unique responsibilities. In short, the law reflects reality: risk depends on usage, not hype. Domain-specific AI acknowledges this reality. It narrows inputs and outputs, aligns evaluations with workflow, and establishes error budgets that correspond to the potential harm involved.
General benchmarks support the same idea. When researchers adapted the traditional MMLU exam into MMLU-Pro, top models experienced a 16-33% drop in accuracy. Performance on today’s leaderboards tends to falter when faced with real-life scenarios. This serves as a warning against unscoped deployment. Meanwhile, government labs now publish pre-deployment safety evaluations because jailbreak vulnerabilities still occur. The UK’s AI Safety Institute has outlined methods for measuring attack success and correctness; its May 2024 update bluntly addresses models’ dangerous capabilities and the need for contextual testing. If failure modes vary by context, safety must also be context-dependent. Domain-specific AI enables this possibility.
The curve for incidents is steep. The Stanford AI Index 2025 reported 233 incidents in 2024, marking the highest number on record. This figure is a system-wide measure rather than a reaction to media coverage. Additionally, the proportion of “restricted” tokens in common web data rose from about 5-7% in 2019 to 20-33% by 2023, altering both model inputs and behavior. As training data becomes more varied, the likelihood of general models misfiring in edge cases increases. Safety cannot be a once-and-done action; it must be a continuous practice tied to specific domains, using ongoing tests that reflect real tasks.
Safety evaluations confirm this trend. New assessment frameworks like HELM-Safety and AIR-Bench 2024 reveal that measured safety is heavily influenced by the specific harms tested and how prompts are structured. The UK AISI’s method also rates attack success rates and underscores that risk is contingent on deployment context rather than solely on model capability. The conclusion is clear: a general score does not guarantee safety for a specific classroom workflow, exam proctor, or FX-desk summary bot. Domain-specific AI allows us to select relevant benchmarks, limit the input space, and establish refusal criteria that reflect the stakes involved.
Data presents another limitation. As access to high-quality text decreases, developers increasingly rely on synthetic data and smaller, curated datasets. This raises the risk of overfitting and makes distribution drift more likely. Therefore, targeted curation gains importance. A model for history essay feedback should include exemplars of rubrics, standard errors, and grade-level writing—not millions of random web pages. A model for news reading in FX requires calendars, policy documents, and specific press releases, rather than a general assortment of internet text. In both cases, domain-specific AI addresses the data issue by defining what “good coverage” means.
In education, the risks are personal and honest. AI tools that determine student access or evaluate performance fall into high-risk categories in the EU. This requires rigorous documentation, monitoring, and human oversight. It also calls for choosing the right design for the task. A model that accepts inputs aligned with rubrics, generates structured feedback, and cites authoritative sources is easier to audit and remains within an error budget compared to a general chat model that can veer off-topic. NIST’s AI Risk Management Framework and its 2024 Generative AI profile provide practical guidance: govern, map, measure, manage—applied to specific use cases. Schools can utilize these tools to define what “good” means for various needs, such as formative feedback, plagiarism checks, or support accommodations, and to determine when AI should not be applied.
In finance, the most effective approaches are focused and testable. The best outcomes currently emerge from news-to-signal pipelines rather than general conversation agents. A study by the ECB this summer found that using a large model to analyze two pages of commentary in PMI releases significantly enhanced GDP forecasts. Academic and industry research shows similar improvements when models extract structured signals from curated news instead of trying to act like investors. One EMNLP 2024 study showed that fine-tuning LLMs on newsflow yielded return signals that surpassed conventional sentiment scores in out-of-sample tests. A 2024 EUR/USD study combined LLM-derived text features with market data, reducing MAE by 10.7% and RMSE by 9.6% compared to the best existing baseline. This demonstrates domain-specific AI in action: focused inputs, clear targets, and straightforward validation.
The governance layer must align with this technical focus. ISO/IEC 42001:2023, the first AI management system standard, provides organizations with a way to integrate safety into daily operations: defining roles, establishing controls, implementing monitoring, and improving processes. Combining this with the EU AI Act’s risk tiers and NIST’s RMF creates a coherent protocol for schools, ministries, and finance firms. Start small. Measure what truly matters. Prove it. Domain-specific AI is not a retreat from innovation; it is how innovation thrives amid real-world challenges.
Critics may argue that “narrow” models will lag behind general models that continue to scale. However, the issue isn’t about measuring intelligence; it’s about being fit for purpose. When researchers modify tests or change prompts, general models struggle. MMLU-Pro’s 16-33 point decline illustrates that today’s apparent mastery could collapse under shifting distributions. General models remain prone to jailbreak vulnerabilities, so safety teams continue to publish defenses because clever attacks still work. The UK AI Safety Institute’s methodologies and the follow-up reports from labs and security firms emphasize one need: we must evaluate safety against the specific hazards of each task. Domain-specific AI achieves this inherently.
Cost is another concern. Building smaller, focused systems may be redundant. In reality, effective stacks combine both types. Use a general model for drafting or retrieval, then run outputs through a policy engine that checks for rubric compliance, data lineage, and refusal rules. ISO 42001 and NIST RMF guide teams in this process by detailing what data was utilized, what tests were conducted, and how failures are addressed. The EU AI Act rewards such design with clearer paths to compliance, particularly for high-risk educational applications and for general models used in regulated environments. The lesson from recent evaluations is evident: governance must reside where the work occurs. This is more cost-effective than dealing with incident responses, reputational damage, and rework after audits.
A final objection is that specialization could limit creativity. Evidence from the finance sector suggests otherwise. The most reliable improvements currently come from models that analyze specific news to generate precise signals, validated against clear benchmarks. The ECB example demonstrated how small amounts of focused text improved projections, while the EUR/USD study outperformed baselines using task-specific features. Industry research indicates that fine-tuned newsflows perform better than generic sentiment analysis. None of these systems “thinks like a fund manager.” They excel at one function, making them easier to evaluate when they falter. In education, the parallel is clear: assist teachers in providing rubric-based feedback, highlight patterns of misunderstanding, and reserve critical judgment for humans. This approach keeps the helpful tool and mitigates harm.
The evidence is compelling. Incidents rose to 233 in 2024, public trust is fragile, and stricter benchmarks reveal delicate performance. The solution isn't abstract “alignment” with universal values implemented in larger models. The remedy is to connect capability with context. Domain-specific AI narrows inputs and outputs, utilizes curated data, and demonstrates effectiveness through relevant tests. It establishes concrete governance through ISO 42001 and NIST RMF. It aligns with the EU AI Act’s risk-driven framework. Its efficacy is already demonstrated in lesson feedback and news-based macro signals. The call to action is straightforward. Schools and ministries should set domain-specific error budgets, adopt risk frameworks, and choose systems that prove their reliability in domain tests before they interact with students. Financial firms should define assistants to focus on information extraction and scoring, not decision-making, and hold them to measurable performance standards. We can continue to chase broad applications and hope for safety, or we can develop domain-specific AI that meets the standards our classrooms and markets require.
The views expressed in this article are those of the author(s) and do not necessarily reflect the official position of the Swiss Institute of Artificial Intelligence (SIAI) or its affiliates.
Adalovelace Institute. (2024). Under the radar? London: Ada Lovelace Institute.
Carriero, A., Clark, T., & Pettenuzzo, D. (2024). Macroeconomic Forecasting with Large Language Models (slides). Boston College.
Ding, H., Zhao, X., Jiang, Z., Abdullah, S. N., & Dewi, D. A. (2024). EUR/USD Exchange Rate Forecasting Based on Information Fusion with LLMs and Deep Learning. arXiv:2408.13214.
European Commission. (2024). EU AI Act enters into force. Brussels.
Guo, T., & Hauptmann, E. (2024). Fine-Tuning Large Language Models for Stock Return Prediction Using Newsflow. In EMNLP Industry Track. ACL Anthology.
ISO. (2023). ISO/IEC 42001:2023—Artificial intelligence management system. Geneva: International Organization for Standardization.
NIST. (2023). Artificial Intelligence Risk Management Framework (AI RMF 1.0). Gaithersburg, MD.
NIST. (2024). AI RMF: Generative AI Profile (NIST AI 600-1). Gaithersburg, MD.
RAND Corporation. (2024). Risk-Based AI Regulation: A Primer on the EU AI Act. Santa Monica, CA.
Reuters. (2025, June 26). ECB economists improve GDP forecasting with ChatGPT.
Stanford CRFM. (2024). HELM-Safety. Stanford University.
Stanford HAI. (2025). AI Index Report 2025—Responsible AI. Stanford University.
UK AI Safety Institute (AISI). (2024). Approach to evaluations & Advanced AI evaluations—May update. London: DSIT & AISI.
Wang, Y., Ma, X., Zhang, G., et al. (2024). MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark. NeurIPS Datasets & Benchmarks (Spotlight).
Takaichi’s rise is a pragmatic bet, not a protest vote Shrinking cohorts demand an education reset—protect funding, cut teacher overload, use AI with guardrails Judge success by delivery: fair consolidation and measurable gains in teacher time
Firms under competition cut jobs, switch lines, or move tasks offshore Schools need a skills policy for competition: fast, stackable, portable training Fund future-shaping sectors and track outcomes so workers can switch quickly
<Physical and transition climate risks are orthogonal inside firms Physical risk is geographic and weather-driven; transition risk is policy- and profit-driven Teach and regulate with two dashboards, not one, so budgets and actions match each risk
AI is making labor borderless as online services surge Opportunity expands, but standards, audits, and broadband are crucial Schools must teach task-first skills, platform literacy, and safeguards
The fastest-growing part of global trade is undergoing a profound transformation. In 2023, digitally deliverable services—work sent and received through networks—reached about $4.5 trillion. For the first time, developing economies surpassed $1 trillion in exports. This accounts for more than half of all services trade. Borders still exist, but for more tasks, they function like dimmer switches rather than walls. This is the world of borderless AI labor, a transformative force where millions of people annotate data, moderate content, build software, translate, advise, and teach from anywhere. The rise of large language models and affordable coordination tools has accelerated this shift. Employers can now swiftly create distributed teams, manage them with algorithmic systems, and pay them across time zones. The key question is not whether jobs will shift, but who sets the rules for that shift and how schools and governments prepare people to succeed in it.
The numbers clearly illustrate this trend. In a 2025 survey of employers, 40-41% of firms indicated they would reduce staff where AI automates tasks. At the same time, they are reorienting their business models around AI and hiring for new skills. This change does not spell disaster. By 2030, the best estimate is a net gain of 78 million jobs globally, with 170 million created and 92 million displaced. To put it another way: tasks are moving, talent strategies are changing, and opportunities are opening—often across borders. The borderless AI labor market is not just a shift, but a gateway to unprecedented growth and opportunity.
Hiring practices are already reflecting this shift. Data from platforms show that remote and cross-border hiring increased through 2024. Over 80% of roles on one global platform were classified as remote, and cross-border hiring rose by about 42% year-over-year. In the United States alone, 64 million people—or 38% of the workforce—freelanced in 2023, many serving clients they never met in person. These trends, coupled with the surge in digitally delivered services, confirm that borderless AI labor has scale and momentum.
The benefits are significant. For the 2 billion people in informal work—over 60% of workers in much of the Global South—AI-enabled matching, translation, and reputation tools can help clarify skills for employers and create opportunities for higher-value tasks. This is the inclusive promise of borderless AI labor, where visibility and trust are essential in global services. When communication, payments, and rating systems are tied to the worker, location becomes less critical, and performance issues become more significant.
However, the risks are real and immediate. The global supply chain for AI, which relies on data labelers and content moderators in cities like Nairobi, Manila, Accra, Bogotá, and others, is a key part of this issue. Over the past two years, research and litigation have highlighted issues like psychological harm, low pay, and unclear practices within this ecosystem. New projects by policy groups are beginning to show how moderation systems can fail in non-English contexts and low-resource languages. The solution is not to withdraw from global work but to set enforceable standards for wages, safety, and mental health support wherever moderation and annotation take place. While automation can filter some content, it cannot replace the need for decent labor conditions.
A subtler risk comes from algorithmic management itself. This refers to the use of AI tools to assign tasks, monitor productivity, and evaluate workers at scale. Recent surveys in advanced economies reveal widespread use of these systems and varied governance. Many companies have policies, but many workers still express concerns about transparency and fairness. In cross-border teams, power imbalances can worsen if ratings and algorithm-based decision-making are unaccountable. The solution is not to ban these tools; it is to make them auditable and explainable, providing clear recourse for workers when mistakes happen.
There is also a broader change. As tasks move online, the demand for office space is shifting unevenly. In the U.S., office vacancies reached around 20% by late 2024, the highest level in decades, with distress in the sector nearing $52 billion. This does not mean cities will die; rather, it indicates a shift away from low-performing spaces toward digital infrastructure and talent. Employers will continue to focus on time zones, skills, and cost—rather than postal codes. Borderless AI labor will keep driving this change.
First, establish portable labor standards for digital work. Governments that buy AI services should require contractors—and their subcontracted platforms—to meet basic conditions, including living-wage floors based on purchasing power, mandated rest periods, and access to mental health support for high-risk jobs like content moderation. International cooperation can help align these standards with existing due diligence laws. A credible standard would connect buyers in cities like New York or Berlin to the outcomes for workers in places like Nairobi or Cebu.
Second, include algorithmic management in policy discussions. Regulators should require companies that use automated assignment, monitoring, or evaluation to disclose such practices, document their effects, and provide appeals processes. Evidence shows that companies already report governance measures, but their implementation often outpaces oversight. Clear rules, alongside audits by independent organizations, can reduce bias, limit intrusive monitoring, and protect the integrity of cross-border teams. Where national laws do not apply, public and institutional buyers can enforce contract obligations.
Third, invest in connectivity. An estimated 3 billion people still lack internet access. A rigorous estimate from 2023 places the global broadband investment needed for near-universal access at about $418 billion, mostly in emerging markets. This investment is essential infrastructure for borderless AI labor. Prioritize open-access fiber, shared 4G/5G in rural areas, and community networks. Public funds should promote open competition and fair-use rules so smaller firms and schools can benefit.
Fourth, speed up cross-border payments and credentials. Workers need quick and low-cost methods to get paid, demonstrate their skills, and maintain verified work histories. Trade organizations are already tracking a surge in digital services. The next step is mutual recognition of micro-credentials and compatible identity standards. Reliable certificates and verifiable portfolios allow employers to assess workers based on evidence, not stereotypes—an essential principle in borderless AI labor.
Education systems must shift from a geography-focused model to a task-focused model. Start with language-aware and AI-aware curricula. Students should learn how to use model-assisted tools for translation, summarization, coding, and data cleaning. They also need to judge outputs, cite sources, and respect privacy. Employer surveys indicate substantial investment in AI skills and ongoing job redesign. Schools should reflect this reality with projects that involve actual work for external clients and peers in different time zones, building the trust that borderless AI labor values.
Next, teach platform literacy and how to promote portfolios. Many graduates will have careers that combine organizational roles with platform work. They must understand service-level agreements, dispute processes, rating systems, and the basics of tax compliance across borders. Capstone projects should culminate in public portfolios featuring versioned code, reproducible analyses, and evidence of collaboration across cultures and time zones. Credentialing should be modular and stackable, so a learner in Accra can combine a local certificate with a global micro-credential recognized by the same employer.
Finally, prioritize well-being and ethics. The data work behind AI can be demanding. Students who choose annotation, moderation, or risk analysis need training to handle exposure to harmful content and find pathways to safer, higher-value roles. Programs should normalize mental health support, rotate students out of traumatic tasks, and teach how to create safer processes—such as automated filtering, triage, and escalation—to reduce harm. The goal is not to deter students but to empower them and provide safeguards in a market that will continue to grow.
The world has already crossed a significant threshold. When digitally deliverable services surpass $4.5 trillion and developing economies exceed $1 trillion in exports, we aren’t discussing a future trend; we’re managing a current reality. Borderless AI labor is not just a concept; it involves a daily flow of tasks, talent, and trust—organized by algorithms, traded in global markets, and facilitated over networks. We can let this system develop by default, with weak protections, inadequate training, and increasing disparities. Or we can take action: establish portable standards for digital work, regulate algorithmic management, invest in connectivity, and teach for a task-driven environment. If we choose the latter, the potential is genuine. Millions who have been invisible to global employers can gain recognition, verifiable reputations, and access to safer, better-paying jobs. This is a policy decision—one that schools, ministries, and buyers can act on now, while the new market is still taking shape.
The views expressed in this article are those of the author(s) and do not necessarily reflect the official position of the Swiss Institute of Artificial Intelligence (SIAI) or its affiliates.
Amnesty International. (2025, April 3). Kenya: Meta can be sued in Kenya for role in Ethiopia conflict.
Brookings Institution. (2025, October 7). Reimagining the future of data and AI labor in the Global South.
Center for Democracy & Technology (CDT). (2024, January 30). Investigating content moderation systems in the Global South.
Deel. (2025, April 13). Global hiring trends show cross-border growth and 82% remote roles.
MSCI Real Assets. (2025, July 15). The office-market recovery is here, just not everywhere.
Moody’s CRE. (2025, January 3). Q4 2024 preliminary trend announcement: Office vacancy hits 20.4%.
OEC D. (2024, Mar.). Using AI in the workplace.
OECD. (2025, February 6). Algorithmic management in the workplace.
Oughton, E., Amaglobeli, D., & Moszoro, M. (2023). What would it cost to connect the unconnected? arXiv.
United Nations Conference on Trade and Development (UNCTAD) via WAM. (2024, December 7). Developing economies surpass $1 trillion in digitally deliverable services; global total $4.5 trillion.
Upwork. (2023, December 12). Freelance Forward 2023.
World Economic Forum. (2025, January 7). Future of Jobs Report 2025 (press release and digest).
World Economic Forum. (2025, May 13). How AI is reshaping the future of informal work in the Global South.
World Trade Organization. (2025). Digitally Delivered Services Trade Dataset.
AI tongue diagnosis turns a centuries-old check into fast, cheap triage With phones and clear protocols, it flags likely risks for confirmatory tests Deploy in primary care with consent, calibration, and monitoring to scale safely
Older adults lose billions to AI elder fraud each year Biology and deepfake tools amplify impersonation and urgency Add default friction—holds, verification, and reimbursements—to stop wires before money leaves
In 2024, Americans
