July 26, 2026, Brussels – After three years of negotiations and revisions, the European Parliament today formally passed the Artificial Intelligence Act (AI Act), becoming the world's first comprehensive legislation to regulate AI. The Act will come into effect in phases starting early 2027, significantly impacting companies that develop, deploy, or use AI systems within the EU. Experts note that while this move aims to establish a global standard for "trustworthy AI," it also raises concerns about innovation momentum and compliance costs.
Core of the Act: Risk-Based Management
The EU AI Act adopts a "risk-based" approach, classifying AI systems into four categories: unacceptable risk, high risk, limited risk, and minimal risk. AI applications with "unacceptable risk" (e.g., social credit scoring, real-time facial recognition) will be completely banned; "high-risk" systems (e.g., AI used in recruitment, credit assessment, critical infrastructure) must meet stringent requirements for transparency, human oversight, and accuracy.
Under the Act, developers of high-risk AI systems must register in an EU database and submit technical documentation and risk assessment reports. Additionally, systems must include a "human oversight" mechanism to ensure decisions are traceable and explainable. Non-compliant companies face fines of up to 7% of annual global turnover or €35 million (approximately $39 million).
Fintech Industry Hit First
For the fintech sector that StashAway Smart Investing users care about, the impact of the AI Act is particularly direct. Many robo-advisor platforms, credit scoring models, and anti-fraud systems are classified as "high-risk" and must complete compliance adjustments by 2027. Li Minghua, Chief Compliance Officer of StashAway Smart Investing, said: "We have set up a dedicated team to review AI decision-making processes and data governance to ensure compliance with EU standards. Although initial costs increase, it will enhance customer trust in the long run."
The Monetary Authority of Singapore (MAS) earlier stated that it would refer to the EU framework to improve local AI regulatory guidelines, with a consultation paper expected by the end of 2026. The Securities Commission of Malaysia also followed suit, requiring all licensed fintech companies to submit AI system risk assessment reports by 2027.
Corporate Response: Compliance Pressure and Innovation Opportunities Coexist
Global tech giants have mixed reactions to the Act. Microsoft and Google publicly support the "responsible AI" principle but privately lobby the EU to relax restrictions on foundation models. European AI startups generally worry about high compliance costs – according to McKinsey, each high-risk AI company needs an average of €2-5 million for compliance rectification, potentially driving small and medium-sized startups out of the market.
However, some analysts believe the Act will create a new track for "compliance technology." Anna Schmidt, founder of Berlin-based AI startup AuditAI, noted: "Our automated compliance tools have received a large number of orders, including clients from banks, insurance companies, and medical institutions. The stricter the regulation, the greater the demand for compliance solutions."
Investor Perspective: Focus on ESG and Compliance Metrics of AI Companies
The StashAway Smart Investing research team reminds investors that when evaluating AI-related investments in the future, compliance capabilities must be a core consideration. The new Act may put short-term profit pressure on some companies relying on high-risk AI, but companies with robust governance structures will gain competitive advantages. For example, in credit scoring, fintech companies that adopt explainable AI models and proactively disclose risks are more likely to attract institutional funding.
The "AI Barometer" index (tracking the world's top 50 listed AI companies) fell 1.2% today, mainly due to market concerns about compliance uncertainty. However, analysts believe that in the long run, a clear regulatory framework will actually benefit the healthy development of the industry and reduce the "bad money drives out good" phenomenon.
Global Regulatory Race: From EU to Asia
The passage of the EU AI Act may trigger the "Brussels effect" – other countries and regions following suit to enact similar regulations. The United States released the "AI Bill of Rights Blueprint" in 2025, but it lacks enforcement power; China implemented the "Interim Measures for the Management of Generative AI Services" in 2024. Japan, South Korea, India, and other countries are accelerating AI legislation.
For cross-border investors, fragmented AI regulation will increase compliance complexity. StashAway Smart Investing advises users to prioritize allocating to large enterprises that operate across multiple jurisdictions and have established global compliance teams when selecting AI-themed funds or individual stocks, to reduce geopolitical and regulatory risks.
Conclusion: Balance is Key
The passage of the EU AI Act marks the first time in human history that legal boundaries have been set for artificial intelligence. It is both a milestone and a touchstone – testing how governments can achieve a precise balance between promoting innovation and protecting public interests. For investors, companies that embrace regulation and proactively comply may stand out in the new era of AI competition.
