US Secretary of State Antony Blinken speaks during "A Conversation on Artificial Intelligence" at the US State Department in Washington, DC, June 28, 2024. ROBERTO SCHMIDT/AFP via Getty Images
On August 1, 2024, the European Union's groundbreaking artificial intelligence law, known as the AI Act, officially came into effect. Per CNBC, this landmark legislation aims to regulate the development, deployment, and use of AI technologies within the EU.
The law was finalized in May after four years of negotiations and is now looking to change how AI will be governed globally, with far-reaching implications for major U.S. tech companies like Microsoft, Google, Amazon, Apple, and Meta.
The AI Act introduces a comprehensive regulatory framework that applies a risk-based approach to AI applications. It distinguishes between low, medium, and high-risk AI systems, imposing stricter regulations on those deemed high-risk.
These include systems used in autonomous vehicles, medical devices, loan decisioning, educational scoring, and remote biometric identification. Companies operating in these areas will be required to implement rigorous risk assessment and mitigation strategies, use high-quality training data, and provide detailed documentation to regulatory authorities.
In addition to regulating high-risk AI, the Act also imposes a blanket ban on "unacceptable" AI applications, such as social scoring systems, predictive policing, and emotional recognition technologies in sensitive settings like workplaces and schools. This part of the law hopes to address ethical issues and possible misuse of AI.
For U.S. tech giants, the AI Act presents a new set of challenges and compliance requirements. The law's extraterritorial scope means that any company with operations or business interests in the EU must adhere to its rules, regardless of where the company is headquartered.
Penalties for non-compliance with the AI Act will likely range from 35 million euros ($41 million) or 7% of global annual revenues, whichever is higher.
The enforcement of the Act will be overseen by the newly established European AI Office, which will monitor compliance and issue fines for breaches.
While the law's full provisions will not be enforced until 2026, including restrictions on general-purpose AI systems like OpenAI's GPT and Google's Gemini, companies are granted a transition period to align their systems with the new regulations.
Join the Conversation